ADVERTISING

advertising as we know it is a development substantially of 75 years, beginning about the mid- dle of the roth century. however, advertising has existed in one form or another as far back as there has been any method of recording ideas by means of visual symbols. i. a historical sketch speaking broadly, the development of advertising may be traced through four stages of growth: (1) pre-printing period prior to 1450. (2) early printing period from 1450 to about 1850. (3) modern period of expansion from 1850 to 1911. (4) the period of development of standards of practice and the introduc- tion of research methods from 1g11 to the present. pre-printing stage-—some form of advertising has probably existed since the time when men lived in communities and com- peted with one another for the necessities of life. advertising on a large scale was of course impossible without a convenient, rapid method of reproducing visual symbols such as was made possible by the invention of the printing press. prior to that time there were forms of handwritten and inscribed announce- ments corresponding to advertisements. there were inscriptions on walls and announcements on sheets of papyrus. perhaps the adulteration—advertising oldest known advertisement of this type isin the british museum, and was written on a sheet of papyrus found in the ruins of ancient thebes, in egypt. this announcement offered a reward for a runaway slave. it was written possibly about 5000 z.c. in rome, during the time of the caesars, there were similar bulletins written by slaves and displayed on the boards erected for that purpose about the city. in both greece and rome ad- vertisements of gladiatorial exhibitions were posted, giving an appearance probably not unlike that of a modern town when it is filled with circus posters. there was an early interest in the de- velopment of trade-marks but only in a very limited manner. these trade-marks were put on vases, pieces of crockery and the like. one form of attracting attention, in some respects a fore- runner of advertising, was that employed by the carthaginians, who, when they arrived with a cargo of goods on the } hoenician shores, built a fire to attract the attention of the inhabitants, who then gathered to examine and purchase the goods. at first the merchant cried his wares on the streets, carrying them with him as he did so. later, when shops were established, the professional town-crier appeared on the scene. ‘there were during the pre-printing period many town-criers in the various larger citics of europe. while their method of announcing or selling was oral, it was nevertheless a forerunner of present-day advertising in the sense of mass selling. it is stated that in 1641 there were 400 town-cricrs in [aris. the early printing stage-—the second stage arose when mod- ern methods of printing were invented. this made possible the rapid and unlimited duplication of writing by means of print which is the necessary medium of advertising. with this devel- opment there came about the publication of newspapers and periodicals corresponding to our modern magazines. it is said that the first newspaper was published in strasbourg in 1609. in america, the first newspaper was entitled publick occurrences both foreign and domestick, published in boston in 1690. this was succeeded in 1704 by the boston news leiter, a weekly pub- lication. the first issue of the boston neus leticr, published april 26 1704, contained advertisements. it was 40 years before this publication reached a circulation of 300 copics per issue. the early history of advertising, together with numerous sample advertisements of this carly period, is well presented in samp- son’s ilistory of advertising. the alodern period.—this period may be dated from 1850 to tot1. the reason for putting the dividing point between the second and third periods at 1850 is the rapid appearance of news- papers and magazines, which made possible the development of modern advertising on a large scale. advertising was impossible until printing developed and until people generally learned to read. one reason for the rapid increase in the number and dis- tribution of advertising media at this time is probably to be found in the development of transportation systems— railways and waterways. prior to that time railway lines were limited to restricted areas. there was no national transportation system. to illustrate the difficulties of transportation before that time, and even up to 1870 or 1875, we may quote the following an- nouncement of the kenosha (wisconsin) telegraph in rowell’s directory in 1869:— the town is renowned for the manufacture of wagons, which find a market all the way to the rocky mountains, and even to oregon, being shipped by way of new york. this shows how difficult it was to distribute goods generally over the country. it is obvious, then, that without the possi- bility of transporting goods there was no need of advertis- ing media of wide circulation, and furthermore media could not develop a wide circulation because they could not be dis- tributed conveniently or rapidly. during the middle and latter part of the roth century the development of a national trans- portation system made possible the ready distribution of goods and publications. another reason for placing the dividing point between the second and third periods at 1850 is that about this time advertising agencies arose whose primary purpose was to serve as brokers of advertising space. they were rather different in this respect from the present advertising agency, which, in advertising addition to contracting for space, is, in its best form, an organi- sation giving expert counsel and service regarding marketing and advertising plans and methods. during this period the number of magazines and newspapers grew very rapidly, so that in 1861 there were 5,203 magazines, papers and periodicals of all kinds in the united states. yet advertising was relatively limited as compared with the present day. harper's magazine had its first advertisement in 1864; scribner’s in 1872. magazine advertising on a large scale did not begin until about 1860 to 1870. it was stated in 1926 that j/ar- per’s magazine had approximately as many pages of advertising in one year as it had in its first 24 years. at first it was used chiefly as a medium for advertising the harper publications. advertising rates at that time began to assume real proportions. harper’s weekly was receiving $35 an inch for its back page in the ’sixties. the first advertising agency was established in philadelphia in 1840 by volney b. palmer. later, offices were established in boston and new york. rates and contracts were unstandard- ised at that time. the agent contracted for a certain amount of space with a publication and then sold it for whatever he could, receiving sometimes as high as 50% of the cost of the space as his commission. among the better known early advertising agents were v. b. palmer, s. m. pettingill and george p. rowell. in 1926 the united states had about 1,200 advertising agencies. some 140, including most of the larger ones, were members of the american association of advertising agencies. development of higher standards and practices -——the fourth period in the development of advertising may be said to have begun in ro1r. since that time advertising has developed most rapidly as a standardised business. the date 1911 is chosen be- cause it was in that year that definite organised steps were taken to forward the movement of truth in advertising. in 1911 counsel for printers’ ink, a leading advertising journal in the united states, formulated what has become known as ‘‘ the printers’ ink model statute,” which has been adopted in a large number of states. this statute makes dishonest advertising a misde- meanour. it (a) places the responsibility for deception upon the adver- tiser, (6) deals with questions of facts about goods rather than opinions and (c) designates the making of untruthful, deceptive or misleading statements a misdemeanour. up to nov. 1921 this statute had been adopted in 22 states in substantially the form in which it was originally prepared by counsel for printers’ ink, and in r§ states in modified form. in these r5 states the modification which was made in the statute consisted for the most part of the insertion of the words ‘knowingly ” or “‘ with fraudulent intent.”’ it is obvious that this change very considerably weakens the forcefulness of the statute. nevertheless, the adoption of this statute has been a powerful force in bringing deception, misrepresentation and dis- honesty in advertising under legal control. the associated advertising clubs of the world, the national and local adver- tising clubs and other organisations have been active in pushing the adoption of this or similar legislation, which has been a powerful influence toward the remedying of objectionable prac- tices. in addition to state legislation a number of larger cities have ordinances intended to check certain undesirable advertis- ing practices. about 1912, the associated advertising clubs of the world at their annual convention adopted the motto “ truth in adver- tising,’ and established a national vigilance’ committee, now called the national better business commission, with the pur- pose of forwarding higher standards of honesty and reliability in advertising. this committee organised, in connection with the various local clubs throughout the united states, local vigi- lance committees or better business commissions to eliminate objectionable, untruthful advertising. the purpose of these commissions has been to carry on locally in each city active efforts, similar to those of the national commission, toward the elimination of dishonest and objectionable advertising and re- lated business methods. up to 1925, such commissions had been ii established in over 40 cities, including boston, st. louis, cleve- land, indianapolis, kansas city, des moines, minneapolis, st. paul, spokane, portland, ore., and new york city. about this same time the associated advertising clubs of the world undertook definite educational activity in outlining courses of study and discussions to be conducted by the various local clubs, for the purpose of studying advertising methods and of improving them. a distinct service has been rendered in this connection by the several educational activities of the different local organisations. colleges and universities began to give serious attention to the study of advertising by establishing well-defined courses of instruction. facts about circulations—in 1914 there was organised the audit bureau of circulations. this has become the leading agency for securing reliable information about the circulation and distribution of publications. the work of this bureau came about as the result of several organisations which were interested in the endeavour to develop a generally acceptable means of obtaining reliable, standardised statements of circulation. it was felt that the time was ripe for devising such means, and that the advertiser had a right to know what he was buying when he bought space in publications. it was desirable to know not only the total circulation but also something about the various aspects of this circulation, facts which are as important as the statement of the total amount of the circulation. about 1870 the only facts generally available regarding circu- lation were the statements in george p. rowell’s directory; and after that time the ayer directery was the chief source of infor- mation. in order to bring about a somewhat more reliable statement of circulations at that time, mr. rowell added a gold mark to the name of the publication if the publisher made an affidavit concerning his circulation. this was accompanied by an offer of $100 to anyone who could disprove the claim made. in general, the a b c reports attempt to answer three main questions: (a) how large is the circulation? (b}) where does it go? and (c) how was it obtained? the bureau operates by requiring from its publisher members a so-called publisher’s statement every six months. the bureau itself makes an audited report once a year. the audit bureau of circulation became of great value to advertising, because it made available reliable statements of circulation covering many different phases. in case of error, either intentional or unintentional, the bureau established cer- tain rules. in case of error, a publisher or advertiser member of the bureau could ask for a re-audit of any publication, if he so desired. if the re-audit proved that the publisher had misrep- resented the facts the expense was to be borne by the publisher. this fact being then bulletined to the other members of the bureau, the publisher became subject to suspension and expul- sion from the bureau. up to jan. 1926, 184 general magazines and periodicals were members of the bureau, 72 farm publica- lions, 240 business publications and g25 newspapers. in addition to publications, 137 national advertisers, 120 local advertisers and 181 advertising agencies were members. about 1913 there was organised in the united states the association of national advertisers, another important influence in the study and development of advertising practices. like- wise, during that period publishers developed a distinct con- sciousness of censorship of advertising. a considerable number of publishers took a definite stand with regard to the advertising which they would or would not accept. the announced or stated policy might not be fully lived up to in practice in all instances but the best evidence of actual progress made in this regard lies in a comparison of the present issues of leading magazines and newspapers with issues of the early years of the century. the improvement is marked. in addition to the elimination of objec- tionable and doubtful advertising, a considerable number of media went a step further by guaranteeing their advertising in the sense that they would either themselves make good on any misrepresentation or deception which appeared in connection with an advertisement, or would see that the advertiser in ques- tion made good. with most of the high-grade media a policy [2 of guaranteeing is substantially implied, and is almost unneces- sary in view of the commendable care which is exercised in rejecting questionable commodities or irresponsible business houses. in t917 the american association of advertising agencies wasorganised. it contributed materially toward the establish- ment of standards of practices and uniform methods of proce- dure. in 1925 it inaugurated a department of research for studying fundamental problems of advertising and markcting. finally, and perhaps most important, an increasing use of re- search methods developed. at least here and there these methods appear to be thoroughgoing in finding the facts on which market- ing and advertising plans may be based, scientific research and planning taking the place of guesswork. il. developments since 1850 considering the last two periods together the chief develop- ments, in addition to those mentioned, may be designated as follows:— growth in ouantity—there has been a tremendous growth in the number of firms employing advertising in one form or an- other. in 1926 there were about 10,000 firms doing more or less national advertising. the report entitled leading advertisers, prepared each year by the curtis publishing co., mentions 1,302 firms, which spent $10,000 in 1921 in 36 leading national maga- zines. in this list there were 50 firms which spent $264,500 or more in 36 leading national publications in 1921. there has been likewise an enormous growth in the number of media. in 1830, it is estimated that there were approximately 800 magazines and newspapers in the united states, of which some 50 were daily publications. their combined annual circulation was estimated at about 60,000,000 copies, or a circulation of approximately 950,000 copies per issue. in 1830 the total population of the united states was 23,500,000. in 1861 there were 5,203 papers and magazines for a popula- tion of approximately 30,000,000. in 1922 there were 22,353 magazines and newspapers in the united states for a population of about 106,000,000. in other words, there has been an increase of magazines and newspapers since 1861 of approximately 5 times, whereas the population during the same period of time has increased approximately 33 times. since 1830 the population has increased approximately 4} times; the number of magazines and newspapers has increased almost 30 times. a more striking increase has taken place in the total combined circulation. the estimated combined circulation per issue of the magazines was (1925) approximately 50,000,000 copies, and the total combined circulation of the newspapers approximately 45,000,000. thus since 1830 there has been a growth in total combined circulation of all magazines and newspapers from less than 1,000,000 to from 95,000,000 to 100,000,000 per issue. besides the marked increase in the number of media and in the extent of their circu- lation, there has been an equally marked increase in the amount of advertising carried in a given issue of a publication. ‘wo representative newspapers in the united states show an increase in advertising space of 5 times and 13 times respectively from 1850 to 1918. during that period, also, there developed entirely new media of advertising, including street-car cards, posters, direct-mail, motion-picture and radio advertising. street-car advertising did not come into general use until about 1890, when the change was made from horse-drawn cars to electric cars. the mail-order field developed rapidly. not only did mail-order houses spring up in large numbers during the period, but also business firms of all kinds developed a very extensive use of direct-mail material in the form of circulars, catalogues, letters and so on. the large mail-order houses commonly sent out twice a year upwards of 4,000,000 large catalogues in addition to numerous smaller departmental catalogues issued. tigures obtained in 1920 esti- mated the total approximate amount spent in direct-mail adver- tising to be $300,000,000; in 1926 the amount was greater, probably much greater. advance tn quality—during the periods under consideration advertising there was a distinct improvement in the quality of advertising. methods of printing, typography, colour printing, art work and the like, greatly improved. besides the mechanical improve- ments there were likewise distinct advances in such essential qualities as accuracy of statement and descriptions, reliability and truthfulness. increase in variety of uses—a widening of the scope of advertising developed rapidly. there was a marked increase in the variety of advertising used. perhaps the dominating types of advertising which flourished originally—that is, about the middle and latter half of the 19th century—were patent-medi- cine advertising and circus bills and in general much unreliable, dishonest and objectionable advertising. as developed there is hardly a business or type of commodity for which the printed word is not used as a means of selling. banks and financial institutions of high grade turned to advertising, finding it profit- able to employ in a dependable manner the methods which had been found profitable by unreliable financial houses. advertis- ing suffered because of the objectionable features which were so prominent in its earlier development. [for this reason, un- doubtedly, high-grade financial houses and to some extent other types of business were slow to take advantage of the possi- bilities which it offered. high-grade financial houses have a dis- tinct responsibility to the community in guiding the savings of the people into the right channels. the unsound and unethical solicitor of investments has taken advantage of the use of the printed word, and has misdirected untold amounts into unsafe channels. high-grade institutions feel that it is perhaps a duty to use all legitimate sales methods equally effectively, thus guiding the investments of the community into proper channels. public service institutions use advertising to place themselve3 before the public in the proper light, to secure the co-operation of the community and to offer their securities to customers and to others. what is termed institutional or good-will advertising is widely used, not only by banking or similar institutions, but by large firms of all kinds, retail stores ancl manufacturers. causes of one kind or another—local, national and international —use advertising to secure publicity and to gain adherents. churches, charitable institutions, schools, educational endow- ments, public health bureaux, political parties, candidates and even governments find advertising a suitable and dignified means of publicity. in industrial disputes, employers and em- plovees alike use advertising as a means of influencing public opinion, securing sympathy and moral support. trade marks —the development of trade marks is likewise a conspicuous characteristic of the periods mentioned, particularly of the first quarter of the zoth century. a large proportion of the trade marks known throughout the united states and all over the world developed during this time. of the somewhat more than 150,300 valid trade marks registered in the u.s. patent office up to 1922 practically none were used prior to 1870, and only a small handful go back to 1885 or 1890. the development of national and international business made trade marks virtually a necessity; and advertising became an impor- tant means, perhaps the chief means, of making them known to the public. volume of advertising —advertising in 1926 played an impor- tant rdle in the operation of business. it assumed such large proportions that it is difhcult to estimate its magnitude and to calculate its exact place in commercial affairs. we may gain some notion of its immense proportions from the amount of money expended and from the amount of space used each year for printed advertising in the united states. it has been esti- mated that in the neighbourhood of a billion dollars is spent annually for this purpose. approximately $500,000 is spent annually for advertising any one of a score or more of well-known commoditics, as high as $3,c00,coo and $4,000,000 being spent in exceptional cases. or we may gain a concrete idea of the im- mensity of advertising from the cost of space for single adver- tisements in some of the leading media. thus, for example, a page ina well-known monthly magazine in a single issue may cost $350, and the back outside cover $800. a page in a well- advertising known weekly publication would run to $7,000, and the back outside cover $12,000. a page in the large metropolitan news- papers cost as high as $1,400 to $1,500. from still another angle the immensity of advertising is indicated by the fact that from two-thirds to three-fourths of the cost of maintaining a newspaper or a magazine comes from its advertising space. the outsider naturally wonders whether advertising with such high rates for space can really be a profitable aid in selling; or whether it may not be an expensive juxury indulged in by large business concerns. the extent to which advertising plays a jus- tifiable rele in the distribution of goods is an important and intricate problem which must be dealt with in any important consideration of the subject. iii. advertising in business there often occurs in the mind of the layman the thought: is not advertising an economic waste? two common opinions with regard to advertising, which are held by a considerable number of people, are that enormously large sums of money are expended for it, and that much of this expenditure must be an economic waste. rumours are spread about that a page in this magazine or in that newspaper costs so many hundreds or thou- sands of dollars for a single issue. or it is learned that the manu- facturer of a certain breakfast food spends so many thousands or perhaps so many millions of dollars for advertising in one year. the inference at once is that the consumer must pay for it in the form of a considerably higher price than the breakfast food or the soap or the suit of clothing would otherwise cost. these impressions usually do not get at the fundamental functions which efficient advertising performs in the distribution of goods. there are wastes in advertising, but there are wastes every- where in production, in distribution, in personal selling. econ- omy in the distribution side of business has lagged behind, but distinct advances have been made. probably the chicf sources of waste in advertising are:— (1) the lack of proper co-ordination between advertising and the other phases of a business—lack of co-ordination with distribution, personal sales, plans and production—advertising where there is no distribution, and the like. (2) the use of inelficient methods—weak, unconvincing and ineffective appeals, poor presentation, ineffective technique, the use of too large space or too small space due to guessing instead of find- ing the facts on which to build and plan. it has been estimated that if every piece of advertising material were made 10°, more effective it would mean a saving of $100,000,000 to the business interests of the united states. (3) the use of advertising for commodities in circumstances in which returns are highly uncertain or for business ventures which are too hazardous. (4) the excessive use of advertising in over-keen competition. (5) the failure to make all advertising literally truthful and de- pendable. even if 96° or 97% of advertising were in the main trustworthy and meritorious, the remaining 3°5 or 4% would weaken the effectiveness of the rest, (6) the use of advertising for illegitimate business, highly specu- lative investments, financial frauds or for commodities of doubtful value or actually injurious to mankind. improvements are proceeding in all these directions. funda- mentally, these sources of waste are not peculiar to advertising, but due to the waste and inefficiency in business as a whole. the same sources of waste exist in oral selling—making plans on guesses instead of on facts, not -being literally truthful and de- pendable in every instance, a lack of proper co-ordination with other phases of a business, the waste of time on the part of sales- men and excessive personal expenditure. preliminary time studies show that the average salesman uses only about 15°% to 50% of his working time in actual selling. fundamentally these wastes in oral as well as in printed selling are not to be laid upon these two aspects of business, but are due to conditions in busi- ness as a whole, to mistaken judgment, to ignorance and moral weakness. the case against advertising —the actual total amount of money spent on advertising is large. the probable amount spent in the united states in one year is in the neighbourhood of one billion dollars. the question is, who pays for it, and how much 13 different would the price of commodities be if this expenditure were absolutely and entirely eliminated, and in what way would the methods of distribution and the cost of distribution be differ- ent? the critic assumes that this money is an economic waste and that prices are higher because of it. a second argument is that advertising has considerably in- creased the demand for, and the purchase of, luxuries and semi- luxuries, such as cameras, beverages, automobiles, pianos, phonographs, vacuum cleaners and so on. a third argument is that advertising has been responsible to a considerable extent for the increasing and widespread use of packages, cartons, wrappers and labels for a large variety of products. advertising has been responsible to a very large extent for the development of standardised brands and trade-marked articles. these wrappers and containers cost money which sup- posedly 1s not a part of the price paid for commodities bought in bulk. the argument is that all of this expenditure is added to the price of the product. relative cost.—it 1s found that the amount actually spent for advertising specific commodities differs widely from the amount which the average person thinks is spent for advertising specific products. almost every person who makes a general statement that the prices of commodities are higher because of the adver- tising, or that the money expended for advertising is wasted, knows little about the actual facts of the relative amount of money expended for advertising a specific commodity. people quite generally have a greatly exaggerated notion of the amount of money spent for printed advertising. an investigation was made to determine the amount which people think is spent for advertising certain commodities. the results of the investiga- tion showed that people generally over-estimate from four to five times the amount spent for advertising specific products. for example, the amount spent for advertising a package of breakfast food retailing at 15 cents was estimated by this repre- ° sentative group of persons as two cents, whereas the actual amount spent by the manufacturer of this food for advertising a single package was three-tenths of a cent, an over-estimation of about eight times. the average person believes that approxi- mately five times as much money is expended for this purpose as is actually the case. averages of expenditures for advertising by a considerable number of firms in various lines of business show that actual figures are relatively small. figures obtained by the harvard bureau of business research, and based on a considerable num- ber of stores in each case, give the percentages of net sales ex- pended for advertising for the years 1919 and 1920 in several lines of business. the accompanying table furnishes the per- centages. | percentages of net sales expended for advertising, 1919 and 1920, in ketat stores common igure mf general stores 0:2 retail groceries. ; 0-2 187 retail drug stores (1919) 0:7 130 ©retail shoe stores (1919) i-3 397. retail shoe stores (1920) : . 1-9 182 retail jewellery stores (1920). ; . 2-0 190 retail jewellery stores (1921) 2°5 266 department stores (1920) . 2-0 301 department stores (1921) . 2°4 the average expenditure for advertising on the part of retailers is approximately 2% or less. certain types of stores, in partic- ular department stores, expend over that amount, and, on the other hand, a very considerable proportion of retailers spend under that amount. figures supplied by reliable sources show that the amount spent by manufacturers for advertising is from 2% to 3%), and that therefore the expenditure for advertising as a whole is probably not over 4° to 5%. it is therefore evident that while the total expenditure for advertising runs into lar_e figures, the relative expenditure in proportion to the cost of a commodity is small. in other words, if advertising played no part whatever in stimulating the sale of commodities, and if the 14 cost of advertising were therefore directly added to the price of a commodity to the consumer, the price of an article would be almost inappreciably higher than it now is. an article selling at $1.00 would have to be sold at $1.04 or $1.05 if advertising were purely a clead expenditure of money. although the actual expen- diture for advertising is relatively smali—probably not over 4% to 5% in the aggregate—even this expenditure would be a serious waste if it produced no economic or social value. the 1% to 2% spent by retailers, or the 2% to 3% spent by manufacturers, would in many instances serve to make the difference between success and failure. effects of advertisement——in 1925 a national distribution conference was held in washington under the auspices of the chamber of commerce of the united states. the report of the committee appointed to investigate ‘“‘ market analysis—adver- tising and advertising mediums,” gives some information with regard to the point just brought up. the investigation consisted of a survey of articles which had appeared in publications, giving specific data about the effects of advertising on costs and prices to the ultimate consumer; and in addition of a questionnaire which was sent out to a group of national advertisers, requesting specific data showing the effects of their advertising on their costs and prices. the results were tabulated and summarised in part as follows:— from the data available it is hardly possible to make a non- disputative demonstration. nevertheless, we believe that it can safely be said that the use of effective advertising by national dis- tributors as a general rule produces one or more of the following results:— (1) decreases the cost of selling. . (2) lowers the cost of production on account of increased volume. ~ (3) lowers prices to the consumer and thereby raises the standard of living. | (4) asa by-product, aids in the ecucation of the general public. data have been obtained from a number of national advertisers to show the effects advertising has had upon the cost of production and the cost of selling, and prices tothe consumer. [t is realised that changes in business conditions during the past few years and changes in distribution policies within the individual companies make it very difficult to eliminate the variable factors an: to prove the elfects of advertising. the data which have been collected, however, seemed to indicate conclusively that the first three results named above are brought about by advertising. ... in addition to the four results obtained by advertising which have been mentioned above, advertising often has the following beneficial effects for the manufacturer, dealer or consumer:— (1) advertising aids in stabilising production. (2) advertising aids in standardisation and quality. (3) advertising aids in the simplification of products. (elimina- tion of unnecessary items in the line.) (4) advertising helps to prevent fluctuation in prices. summary of advantages.—a beneficial effect of advertising has been the reduction of seasonal] fluctuation in the volume of business. in certain fields of business, the seasonable fluctua- tion, if not entirely eliminated, has been greatly reduced. the two seasons of buying men’s clothing have been greatly extended, and the sale of toys, formerly largely confined to the holiday season, is more evenly distributed throughout the entire year. the same is true of fountain pens, millinery, arms and ammuni- tion, oranges and lemons and other products. the result has been a reduction in the cost of doing business. advertising has aided materially in the standardisation of products and quality of goods. this has been brought about through the establishment of brand names and trade-marks, which have become associated with uniform quantities and qualities in the mind of the public, until the recollection of a brand name of any number of specific products calls up the pic- ture of a certain definite quantity and a certain recognised standard of quality, which the public has been educated to receive when it calls for a product by a specific trade-name. although the establishment of standard packages has probably added to the cost of commodities to some extent, it has had certain other advantages. it has facilitated retailing, requiring less time on the part of assistants in the preparation of products for delivery to the consumer, and it has secured a more hygienic handling of goods, particularly foods. advertising the noteworthy advances made in the general level of living, in public taste, in personal and domestic conveniences, in the general use of cultural agencies, must be attributed in part to the use of the printed word in the form of advertising. such products as phonographs and radios and the development in musical taste generally, vacuum cleaners, better-fitting clothes, better- constructed and designed houses, better-painted houses, evidence the advance made in the level of living, due in a considerable measure to the insistent and continuous force of acvertising. there is no doubt that we should very likely have most if not all of these improvements and conveniences without advertising, but probably not so many people would have had them so soon. printed publicity has no doubt helped to educate people to want more and better things in food, clothing, housing, personal com- fort and enjoyment. the almost universal reading of magazines and newspapers has been made possible largely, if not solely, through advertising, since approximately three-fourths of the income of a periodical is derived from its advertising space. in this sense, the news- paper or magazine is practically a by-product of advertising. the trifling sum which is now charged for a newspaper or maga- zine—in comparison with the large sum which such magazine or newspaper would cost if it were not for the advertising which it carries—has made some form of reading matter available to even the poorest classes of persons. advertising revenue has made possible the large-scale production of newspapers, maga- zines and periodicals of all kinds the social and educational value of which is beyond estimate. it seems to have been found from experience that some prod- ucts may be sold most economically primarily through personal salesmanship, that some products may be sold to certain classes of people most economically primarily through printed sales- manship, but that probably the great majority of products may be distributed most economically by varied proportions of per- sonal and printed selling. so long as competition is believed to be a desirable factor in human welfare, just so long will compet- itive means be used. critics of advertising perhaps may neglect the fact that if advertising were eliminated or abolished other competitive methods would necessarily be substituted for it. if the use of advertising were eliminated, it is quite likely that its place would be taken in most instances by some form of per- sonal selling, with its accompanying expense. the fact that numerous firms which have consistently and continuously used printed salesmanship over a long period of years have survived from early periods of business activity may be taken as some indication that advertising has been found to be a fairly effective agency of business progress. (dy st) iv. british and european advertising the conditions under which advertising is conducted in great britain and other european countries differ in many directions from those prevalent in the united states. for this reason british practice diverges in certain important respects from american practice. both advertising as such, and the newspaper, appeared slightly earlier in great britain than in the united states. the first advertisements published were handbills for the sale of goods, printed by caxton late in the 15th century. the first daily newspaper printed was the london gazette, founded in 166s, which is still published. but nathaniel butter had produced his historic weekly newes in 1622. probably the first trade advertisement to appear in a newspaper was one advertising tea, published in mercurtus politicus in 1658. the great national newspaper, the times, was first published in 1678 as the daily universal register. quite a number of existing provincial newspapers, including the notlingham jour- nal, the bristol times and mirror, and the hereford journal, were founded between 1710 and 1713. fifty years later, dr. johnson was complaining that adver- tisements were “‘ so numerous that they are negligently perused: it is therefore becoming necessary to gain attention by magnifi- cence of promises and by eloquence, sometimes sublime and sometimes pathetick.”’ so early as 1712 a duty was placed on advertising advertising. for many decades it yielded the exchequer little revenue. in the roth century, however, it had become a valuable source of income. the tax amounted at first to 1/- on each aclver- tisement. in 1815 it was raised to 3/6, and in 1833 it was reduced to 1/6. in june 1853, the advertisements carried in the times averaged 1,500 each day, and in one day in june there were 2,250 inserted—the greatest number that had ever appeared in one paper. the duty on advertising was finally repealed in 1853. generally speaking, advertising did not make any great head- way in great britain until about the year 1850. modern adver- tising came into existence about 1890, while the organisation of advertising entered its national phase only in the year 1920. the introduction of research methods dates from about the same year. at every stage of development great britain has tended to follow 10 to 20 years behind america, while other european countries tend to lag behind british development and practice by an even wider margin. between 1920 and 1926 british advertising entered on a period of intensificd activity. in these years standards of practice were improved enormously. the national organisation of advertising was not merely initiated but developed with remarkable energy. expansion of british advertising.—some idea of the expansion of advertising in great britain may be gathered from a few figures based on extensive researches. there were in 1926, 2,150 individual newspapers published in great britain, and roughly 1,500 journals and magazines. whereas in rooo there were scarcely too advertising agencies existing in great britain, no fewer than 500 agencies were in active business in 1926. in the same year 340 national and regional bill-posting contractors were at work; 67 firms engaged in outdoor publicity of other types; 168 firms in sign writing and in window dressing; and almost 400 in the making of cartons and containers. there are over 50 commercial art studios in london and the provinces; 39 com- mercial photographers; and several hundred free-lance commer- cial artists. all these concerns and individuals give service to the advertiser. at the same time, seven concerns and individuals devote their energies to market research—a service which had scarcely been heard of in this country prior to 1920. expansion has been no less remarkable in the colonies. influence of conditions on practice—the divergence between american and british advertising methods may be seen clearly by considering a factor which influences marketing practice in every country, viz., transport. transport finally determines the distribution of the various advertising media. in the united states the distribution and influence of the newspaper is purely local. in great britain the distribution of the london newspaper is national. this makes it possible for the advertiser to broad- cast his message from one end of the country to another in 24 hours. since distances in great britain are comparatively short, and since several of the great london newspapers print simul- taneously in london and a northern city (manchester or leeds), complete national distribution of a paper printed on monday evening is secured by tuesday morning. circulations and adver- tising rates must be correspondingly high. circulation.—the circulation of the best american daily news- papers was in 1926 roughly 200,000 to 600,000; the price of a page advertisement usually ranged from £200 to £300. in the same year the circulation of the daily mail, on the other hand, was over 1,700,000, and the price of the front page {1,400. the times achieved the remarkable revenue of £5,000 per day. one popular sunday newspaper actually exceeded all american cir- culations. the news of the world was bought weekly by over 3,000,000 readers, the page rate being normally £2,200. the magazines of great britain do not represent, therefore, as amer- ican magazines do, the sole mearis of securing national circula- tion, nor have they so large a population to tap. yet the unique humorous journal punch with a circulation of over 200,000 copies, carried over £8,000 worth of advertising every week. good housekeeping, a successful associate of the american journal of that name, had an average monthly revenue of over £12,000, vhile its christmas numbers carried over £18,000 in advertise- ments. i5 mail order business —owing to the compactness of the country and the transport facilities available a second variation occurs. the british public is much less susceptible to mail-order advertising than the american. the spread of motor-bus serv- ices to the remotest hamlets may have had much to do with this. the shop window has been brought within reach of the secluded rural populations. the united states post office made it easy for the mail order advertiser by collecting the amount of the purchase through its postmen. on the other hand the mail order advertiser in great britain, had until 1926, to bear the expense of collection by post, as it was not until march of that year that cash on delivery was introduced. further, facil- ities for securing lists of prospective buyers, graded as to in- come or trade, did not exist in english cities. it was impossible to buy what are known in america as lists of tested and guaran- teed names. nor did the postal authorities assist by charging low rates for the carrying of sales literature. scercity of statistics —two further factors tended to make all advertising in great britain or in other european countries more of a “‘ hit or miss ”’ business than in the united states. the first of these was the absence of statistical data on which to assess the market available for any goods either in terms of what com- petitors are doing or of the consuming public concerned. the british board of trade, in particular, reduced its statistical records to the barest minimum. the taking of a new census of production, which must be the basis of all sound market esti- mates, was delayed by the world war and had had its potential value considerably reduced by the disinclination of manufacturers to give certain essential facts concerning their output. —— the united states govt., on the other hand, made advertising more scientific by presenting to the advertiser the finest sta- tistical material available in any country. when the report by the bureau of efficiency on the statistical work of the united states govt. was completed in 1922 it ran to more than 4oo pages. a condensed report, published later, gave no fewer than 700 subjects on which that government compiled statistics, with details as to where and how to secure the facts. | unless statistics of production are made available annually or bi-annually; unless sales records are obtained from every considerable industry; unless exhaustive researches into the consumer habits are made in every area, it is not possible for the british advertiser to approach the scientific accuracy or the technical skill of the american. finally, the british public, being homogeneous and speaking one tongue, is less accessible to the magic of the printed word. in other words, it 1s more sophisticated. added to this sophistication is a keener buying sense, particularly on the part of women, due chiefly to the pos- session of a shorter purse. the resulting influence of all these factors is remarkable. it must mean too often that an advertising campaign successful in amcrica fails completely in great britain. a reconsideration of the marketing and advertising channels available is necessary, along with a modification of copy, illustrations and methods generally, if advertising is to achieve its purpose in great britain or in any other part of europe. law.—up to 1926 no act had been passed in great britain corresponding to the printers’ ink model statute formed in ro11 by printers’ ink and passed in a large number of american states. on the other hand, no such general statute seemed neces- sary, since many of the interests of the advertiser and of the reader of advertisements were already protected in english law by existing statutes; while a body of case law was being built up by test cases in the king’s bench and chancery divisions of the high court of justice. the indecent advertisements act (1889) made it a punishable offence to exhibit or deliver advertisements of an indecent nature. later rulings confirmed the fact that, under the advertising station (rating) act (1899), ground used for advertising is rateable. legislation was certainly required to define the extent to which the statements in an advertisement may be binding on the advertiser. in an important case (carlill v. the carbolic smoke 16 ball co.) a preventive against influenza was advertised, and the makers stated in print that if anybody using the carbolic smoke ball as directed caught influenza within a certain number of days they would pay him f100. the plaintiff used the remedy, and contracted influenza, winning the sum of {100 as the result of his suit. under english law any statement of a specific fact as to the quality or nature of an article in an advertisement amounts to a warranty, the breach of which gives rise to a right of action. as evidence of this we may take the case of a gunsmith who advertised a gun of first-class quality capable of reasonable wear and tear for several years. when the gun was first used 1t burst and injured the plaintiff, who successfully brought an action against the vendor for fraud in inaccurately describing the gun. the position of the advertiser under the copyright act (1911) was, in 1926, still the subject of tigation. perhaps the most important ruling on the subject was that of lord romilly:— copyright exists in published books, periodicals, pamphlets, prints and designs. if you copy the advertisement of another you do him no wrong, unless in so doing you lead the public to believe that you sell the articles of the person, whose advertisements you copy. wherever the letterpress bears the traces of original composi- tian it is entitled to protection, but not where it simply describes the contents of a warehouse, the exertion of the proprietor or the com- mon mode of using familiar articles. in general it may be said that the pla.ntiff must prove damage in a british court of law in order to prove that a competitor has transgressed the law by imitating an advertisement or design. organisations. —the organisation of british advertising on a national scale was first actively undertaken in the year 1920. the stimulus to action probably lay in the great advertising exhibition held at the white city, london, in that year, which first brought ali the advertising interests together for common action. soon after the holding of this exhibition british adver- tising men began to attend the annual conventions of the asso- ciated advertising clubs of the world in america. as the result of these visits, it was decided to hold the convention of 1924 in london. about this time most of the chief advertising bodies in great britain became members of district 14 of the associ- ated advertising clubs of the world. in jan. 1926, district 14 of the associated advertising clubs of the world became the advertising association of great brit- ain. this body had on its committees representatives from all the advertising interests in the country. under its auspices annual advertising conventions are held. as a result of these conventions, an impulse has been given to every phase of adver- tising in great britain. in particular the club movement has grown up throughout the provincial cities. in 1926 there were over 20 advertising publicity clubs affiliated to the advertising association. these clubs decided to meet together frequently for discussion on technical subjects connected with advertising; they undertake arrangements for the local delegates attending the annual convention of the advertising association, and do much work of an educational character both within their own ranks and among the outside public. 7 probably the advertising club movement in great britain ts the most remarkable instance of craft interest shown by any industry or profession. the clubs included the women’s adver- tising club of london, whose membership was to be conftned to women in executive positions. it is to the clubs that the success of the annual advertising conventions is largely due. in 1920 there came into being the incorporated society of british adver- tisers. this body was designed to safeguard the interests of the advertiser, its general objects being to increase sales at reduced costs of selling; to make advertising a more profitable proposi- tion; to make collective expression and action possible to adver- tisers. the association corresponded exactly with the association of national advertisers of america. following two conferences with representatives of the latter body, it was definitely decided in 1924 that the incorporated society of british advertisers should make its organisation as nearly as possible an exact counterpart of the association of national advertisers of amer- ica. the association represented directly nearly 300 national dy. advertisers, and, through the affiliation of trade organisations, over 2,000 firms which advertise. the association had on file all procurable information regarding upwards of 6,000 publications of interest to national advertisers, it campaigned consistently for the institution in great britain of an audit bureau of circu- lation, but hitherto without immediate success. advertising agencies —advertising agents were represented by the association of british advertising agents, of which nearly all the leading agencies in great britain became members. it was registered as an incorporated company in 1917. the deh- nition of an “ advertising agent ” laid down by the association is as follows: “ accredited advertising agents are independent persons, firms or companies of good repute and training, engaged solely in the advertising agency business and possessing the equipment and organisation necessary for the carrying on and development of advertising.” advertising agents are usually remunerated by a commission on the appropriation, ranging up to 15° according to the size of the appropriation and the amount of work entailed. this commission is usually provided by the special agency commis- sion allowed to recognised advertising agents by newspapers and periodicals. important papers and groups of papers have lists of recognised agents, and do not allow advertising agents’ com- mission to firms who have not first been placed on their list. the policy of the association of british advertising agents was direct- ed to encourage its members to improve the service which they were able to offer to their clients and to make the minimum agency remuneration 10%, below which figure the association did not consider it possible for an advertising agency to render adequate service to its clients. the scapa society—possibly the most characteristically british movement associated with advertising was one to restrain its activities. this was the scapa society, formed in 1893 as “the society for checking the abuses of public advertising.”’ the association was concerned with advertising only where the display injuriously affected the amenities of outdoor life. in 1907 the society secured the passage of the advertisements regulation act, and in 1925 of a supplementary act. the © original act authorised local authorities to make bylaws for the control of hoarding over 12 ft. high, and for the regulation or prohibition of advertisements affecting the amenities of a public park, pleasure promenade or landscape. the extended powers included some for the protection of the view of rural scenery from a railway or highway, the amenities of any village within the district of a rural district council, and those of any historic or public building or monument. a remarkable evidence of a fine spirit on the part of the adver- tising profession was the way in which great advertisements were withdrawn by mutual agreement between the scapa and those concerned. the association virtually abolished the sky sign. in 1923 three great oil and tyre concerns removed their road- side advertisements in deference to the desires of scapa. the fraudulent advertiser is dealt with by a national vigilance com- mittee, under the auspices of the advertising association. (cac) ady, endre (1877-1919), hungarian poet, was born nov. 21 1877 at ermindszent, transylvania. he was a leader of the modern school and one of the greatest lyrical poets of hungary. during his youth he lived in the country and became intimate with the publicists and politicians of the radical and socialist parties. when he settled down in budapest, he turned his atten- tion toward revolutionary politics and contributed to several radical papers. he also made prolonged visits to paris and the riviera but returned finally to hungary owing to ill-health, and lived there quietly until his death on jan. 27 1919. during the world war he adopted a pacificist attitude and prophesied the karolyi revolution of 1918. during his lifetime he published about 20 books, poems, novels and essays, but this number was increased year by year by the publication of posthumous works. _ brstiocrapiy.—jean horvath, ady (budapest, 1910); paul gulyas, addy (budapest, 1925); l. hatvany, das verwundete land | (ziirich, 1921); articles in the revue de geneve (1922 and 1924). ae—aerial law ae: see russell, g. w. aehrenthal, aloys lexa von, counr (1854-1912), austro-hungarian statesman, was born sept. 27 1854 at gross- skal, bohemia. in 1899 he became ambassador in st. peters- burg, where he remained until his appointment as foreign minister in october 1906. determined to preserve the interests of austria-hungary in the balkans, he also showed himself will- ing to meet the russian wishes on the dardanelles question. after the outbreak of the young turk revolution in 1908, he negotiated, in accordance with his previously conceived policy, with isvolski, an agreement for the annexation of bosnia and hercegovina to austria-hungary and for the opening of the dardanelles to russian warships. isvolski’s proposals were, however, wrecked by the opposition of great britain, and he protested against the annexation of bosnia and iercegovina; aehrenthal, therefore, abandoned the idea of a friendly accommodation with the russian govern- ment. he held, however, with tenacious energy to his purpose with the result that the annexation was acknowledged by the powers; an agreement was reached with turkey; serbia, after long hesitation, was obliged to submit. for this achievement aehrenthal was rewarded by the rank of count (aug. 18 1900). he held fast by the triple alliance, as the surest bulwark of peace, and sought to form the most intimate relations with the german empire although he insisted on the independence of the habs- burg monarchy and energetically repulsed all efforts on the part of the german chanccllery to set limits to that independence. one of his most difficult tasks was to adjust the ever-recurring conflicts with italy, who, while officially supporting the political action of the triple alliance, often embarked on courses directly opposed to the interests of austria-ilungary. a succession of agreements which he concluded with the italian foreign minis- ter, tittoni, justified his eflorts, and enabled him to maintain correct relations with the italian government. yet, by the maintenance of his peace policy, which had the full approval of francis joseph, he came into serious conflict with the party led by the chief of the general staff, conrad von hetzendorf, which championed an energetic, warlike policy. the battle, carried on with tenacious endurance on both sides, ended in 1911 with the victory of aehrenthal and the resignation of hetzendorf. aehrenthal died feb. 17 1912 in vienna. see b. molden, alois, graf aehrenthal: sechs jahre auswertiger politik oesterreich-ungarns (1917); the article ‘ achrenthal”’ in the deutsche nekrologer (vol. 18, 1917, pp. 230 et seg.); a. f. pribram, “der ionthkt conrad—aehrenthal ”' oesterretchische rundschau (august 1920). (a. f. pr.) aerial law.—aerial law must be considered under its two headings, international (subdivided into peace and war) and national the international law of the air in peace is to be found in the (multilateral) air navigation convention of oct. 13 1919, and in the (bilateral) agreements which certain states have con- cluded as between themselves. the international law for war is not the subject of any general convention; the few references to air warfare in the conventions and declarations of the hague of 1899 and 1907 are now obsolete. the national or municipal law of the air is to be found in the statutes, decrees and regulations of the various states. the convention of 1919—the air navigation convention, 1gt9, the present parties to which are the british empire, belgium, bulgaria, czechoslovakia, france, greece, italy, japan, persia, poland, portugal, rumania, the serb-croat- slovene state, siam and uruguay, begins by affirming the sovereignty of each state over the air above its territory. this principle afhrmed, the convention goes on to provide for freedom of innocent air passage over the territory of the contracting states for each party to the convention, provided the conditions laid down in it are observed. such freedom of passage does not apply to “ prohibited areas’ which states may desire for military reasons to close to all private aircraft, including those in their own territory. | the “ nationality ” of an aircraft is made to depend upon the country in which it is registered; to be registered in a particular 17 state the aircraft must belong to a national of that state or toa company of which the chairman and two-thirds of the directors possess that nationality. double registration is prohibited. every aircraft engaged in international navigation must possess a certificate of airworthiness issued or rendered valid by its state of nationality, and the pilot and other members of the crew must be provided with certificates of competency and licences sim- ilarly issued or validated. the documents (certificates, licences, log-books, passenger lists, manifests) which aircraft must be provided with are laid down. foreign aircraft may be required to land at aerodromes fixed by the state whose territory they enter; and each state has the right to reserve in favour of its national aircraft the carriage of persons and goods for hire between any two points in its territory. provision is made for the rendering of assistance to aircraft in landing or in distress, for the application, to aircraft wrecked at sea, of the principles of the maritime law of salvage, for the use of aerodromes at a common tariff of charges, and for certain other matters. the convention is not applicable to military air- craft nor to state non-military aircraft employed on customs and police service; other state aircraft, e.g., postal, are subject to its provisions. a military aircraft is defined as one commanded by a person in military service detailed for the purpose, and such an aircraft may fly over another contracting state’s territory only in virtue of a special authorisation. rules in regard to the marking of aircraft (to show nationality and identity), certificates of airworthiness, log-books, lights and signals, ‘‘ rules of the road,” licensing of personnel, aeronautical maps and ground markings, the collection and dissemination of meteorological information, and customs procedure, are con- tained in annexes to the convention. international commission.—a permanent international com- mission for air navigation, composed of representatives of the contracting states, is established and charged with certain powers, including the amendment of the annexes, the receipt or initiation of proposals to amend the convention itself and the inter-communication of information affecting air navigation, etc. the office of the secretary-general of the commisson is at 20, avenue kleber, paris. the convention may be denounced by one year’s notice. states which did not take part in the world war may adhere to it at any time; those which took part in the war may adhere under certain specified conditions. other arrangements —a number of important powers, it will be seen, are not as yet partics to the convention, and inter- national air trafiic is regulated, so far as they are concerned, by a series of special agreements between themselves and with other states. the terms of these agreements are in general conformity with those of the convention. the obstacle to a more general acceptance of the convention has been, first, the provision therein forbidding contracting states to permit, except by special and temporary authorisation, the flight above their territory of the aircraft of non-contracting states; secondly, the ‘‘ weighted ”’ vote which was allowed to the british empire, france, italy and japan. protocols of amendment of both these provisions have been signed but, not having been ratified by all the parties, are not yet in force. the convention of 1919 does not bind the contracting parties as belligerents or neutrals (though it would continue to apply in war as between two neutrals), and an attempt to frame a cor- responding code for war was made at the hague in 1923. a commission of jurists, representing the united states, great britain, france, italy, japan and the netherlands, under the chairmanship of judge bassett moore, drew up rules which, though they have not been embodied in a convention, are a valuable indication of international opinion, as it stood in 1926, upon the law of air warfare. the municipal legislation which has been passed to regulate internal flying and to give effect to the international obligations of each country under the convention of 1919, or the separate bipartite agreements, is contained in laws, decrees, ordinances and regulations of many different kinds. it is of great volume and any full survey of it would be impossible here. an indication of 18 its general scope and one or two of its more important features is all that can be attempted. national or municipal legislation.—one usually finds in such legislation (aio sensi) provisions for ensuring the safety of air navigation as affecting both air travellers and the general public; provisions, that is to say, for registration and certification of air- craft, licensing of personnel, inspections and tests precedent to such certification and licensing, maintenance and overhaul and renewal of certificates and licences, inspection, licensing and use of aerodromes, carriage of explosives and arms, dropping of articles in flight, jettison of cargo, minimum height of flight over populated areas, investigation of accidents and so on. for defence reasons “ prohibited areas” are usually defined, and in many cases corridors of entry for foreign aircraft are prescribed, as well as the general conditions in which such entry is allowed; and the carriage of radio apparatus and cameras is usually re- stricted. conditions for the grant of concessions for the operation of commercial air lines and the establishment of international airways are occasionally included. penalties (fine, imprisonment, suspension of licence) for the infringement of the rules are invariably laid down. liability for damage.—the important question of liability for damage occasioned by aircraft to persons or property on the ground is legislated for in many of the codes. in the absence of special provision the liability of the owner or pilot would be de- pendent ordinarily on proof of actual negligence or intention. froof would often be very difficult to obtain and hence it has generally been considered equitable to impose upon the owner or pilot an absolute liability to pay compensation for all such damage, save in so far as he can show that the person damnified was himsclf to blame. it is also sometimes provided that the owner of an aircraft, or the organisers of an air navigation com- pany must cover themselves by insurance against possible claims for compensation or else must deposit security that will be suffi- cient to cover their liability. occasionally it is provided that legal relations between parties on board a foreign aircraft in the air are governed by the law of the aircraft’s nationality, 7.¢., the country in which it is regis- tered. the french law qualifies such a provision by adding that a crime or misdemeanour committed on board a foreign aircraft is within the competence of the french courts, if the guilty party is a french national, or if the aircraft lands in france after the occurrence of the offence. (j. m. sp.) conference of 1925.—the international commisston for air navigation, like other specialised organisations of the league of nations in their several spheres, by its debates and by its labours will be able to give form to the new principles of air traflic regulation and to evolve the international aerial law of the future, all the more surely when all the states concerned par- ticipate in this work. in oct. ig25, after only a few years actual experience of inter- national air traffic, at a conference in paris of the european and certain other governments, a draft convention was framed re- garding the responsibility of carriers by air which is the counter- part, so far as concerns air traffic, to the berne convention of 1890 on transport by rail. an international committee of ex- perts is to continue the work of this first conference, and to study a series of questions, connected with unification of private law, of the very greatest practical importance for the development of air traffic, such as damage caused by aircraft to property and persons on the ground, compulsory insurance, establishment of air registers, rights of property and ownership, mortgages in respect of traffic by air, etc. air lawis already grappling with the practical problems which arise owing to the advance of technical science and commercial organisation. there are in existence permanent international bodies to co- ordinate the efforts of the governments concerned and at the same time to keep in touch with the requirements of those engaged in the traffic. a system of law will thus develop progressively by the solution of individual questions, perhaps, without being guided by preconceived principles.’ the true principles will be embodied in specific regulations which, many years hence, when aerial navigation air traffic has acquired a stable economic regime, will appear as the fruit of long and patient international co-operation in air mattlers—a co-operation as yet in its infancy. (r. ha.) binirograruy.—j. if. spaight, air power and war rights (1924)3 r. b. sparkes, the law refating fo the -lir (1925); manuals of inter- national law, e.g., p. fauchille, treatise on public international law, vol. 1, pt. 2 (1925); fearbooks of the institute of international law, vol. 19, 23, 24, 27, 29; reports of the iniernational law association; records of the international congresses of air legislation, organised by the international legal committee on aviation (7th congress lyons, 1925); bulletins of the international air traffic association (the hagut); bul. no. & of the international commission for aur navigation (4925) containing the convention of oct. 13 191g, re- vised text with the new amendments and protocols; bilateral con- ventions between states, especially great britain, france, belgium, the netherlands, switzerland, germany, denmark, sweden and poland; see league of nations treaty series; half-vearly bueletins of the international conimutsston for air navigation (paris); the legal international review on air traffic (paris, 1910-4, 1922, etc.). aerial navigation.—the practical development of the art of flying has reached the stage at which flights of several hundred miles, without landing, are of frequent occurrence. the trans-atlantic crossing made in 1919 with a vickers-vimy acroplane, involved, in fact, a non-stop flight of 1,900 m. entirely over sea. in order that such long-distance flights mav be made practically independent of meteorological conditions and devoid of the fortuitous element associated with pioncer effort, it has been considered important to enquire how far the existing methods and instruments used for maritime navigation may be applied to navigation in the air, and, where necessary, to devise new methods and instruments for aeronautical purposes. the most elementary method of navigation is the recognition, either from maps or from previous experience, of the country over which the aircraft is flying; for this purpose, maps have been made which give prominence to railways, rivers and other objects conspicuous from aircraft. maps are naturally of no assistance when the craft is flying over sea, desert or uncharted country; moreover, their use necessitates visibility of the ground over at least a considerable proportion of the air route. dead reckoning navigation —the basic method of navigating an aircraft is known as dead reckoning (d.r.); other methods, such as the use of wircless beacons or the determination of the altitude of heavenly bodies, serve only to check the d.r. position. dead reckoning necessitates that a record shall be kept of the speed of the aircraft over the ground and the “ track,” the latter being the actual direction along which the aircraft is passing over the ground. these values may be obtained f10m a knowledge of (a) the speed and direction of the wind, derived either by observation from the aircraft or from meteorological reports; (b) the true speed through the air, obtained by correcting the air-speed indicator for height; and (c) the compass course. if a vector triangle be drawn such that one side represents (a) above, and another side represents (5) and (c), then the length of the third side represents the ground speed and its direction represents the track of the aircraft over the ground. sometimes the third side of the triangle is found by taking ob- servations of the ground, without first determining the wind speed and direction, but the latter values are always recorded, since, if they remain constant, the ground speed and the track may be found for any values of the air speed and compass course, without making further observations of ground objects. the ground speed may be determined by noting the time taken for a ground object, viewed through a “ sight ” set at the appro- priate position on a vertical bar calibrated in terms of height, to appear to traverse the distance between two timing beads fixed on a horizontal wire which ts set along the track of the aircraft. the distance between the two timing beads is made to represent a certain distance on the ground (usually 3 m.) over which the aircraft passes in the observed time. the angle between the fore and aft line of the aircraft and the track is called the drift angle; it may be determined by observing the direction in which an object on the ground appears to move with reference to the fore and aft axis of the aircraft. if the aircraft were flying directly up or down wind, the drift angle aerial navigation would be zero. the true air speed being known, the wind speed and direction may be found, without timing a ground object, by observing the drift angle on two different compass courses, preferably separated by a fairly large angle. graphical construction cannot conveniently be employed in the limited accommodation afforded by an aeroplane cockpit and it is much preferable to solve the vector triangle by instru- mental means. moreover, it is but natural that, where possible, the means for observing ground speed and drift, and the mecha- nism for solving the vector triangle should be incorporated in one instrument. instruments and metitods the following are the most important of the instruments used for d.r. navigation. each enables the ground speed to be determined in the manner described above, and incorporates means whercby the vector triangle may be solved on the instru- ment itself. they differ somewhat, however, in the method by which the drift is determined. the aero bearing plate.—in this instrument the drift angle is determined by sighting on a ground object ahead of the aircraft and turning a horizontal drift bar so that the ground object appears to move along the har. “2 the wind gauge bearing plate—this instrument enables the drift angle to be determined by taking tail bearings of objects over which the aeroplane passed some minutes previously. these observations cannot conveniently be taken by the pilot and a navigator must be carried. —_ ; the course-setting sight.—in this instrument the drift angle 1s determined by sighting ahead. the incorporation of a compass and a link mechanism renders the operation of the instrument to some extent automatic. allimeters.—the most convenient method of determining the height of an aircraft above the ground is by the use of a form of aneroid barometer, known as an altimeter, which is calibrated in terms of height. the pressure on the earth’s surface, due to the weight of the atmosphere above it, is well known to be less on a mountain than it is at sea-level. if, at any particular height above sea-level, the atmospheric pressure was invariable, its measurement would be an ideal means of determining height. the variation of the atmospheric pressure at ground level js, however, a matter of common observation; moreover, the de- crease of pressure with increase of height is governed by the temperature distribution of the atmosphere. all altimeters are, therefore, set at zero when the aircraft leaves the ground, and, in future, are to be calibrated in accordance with a standard atmosphere, which has been adopted internationally. the standard atmosphere assumes that, at sea-level, the tempera- ture is 15°c. and that the barometric height, reduced to o°c., is 760 mm. of mercury. the decrease of temperature with in- crease of height is assumed to be6-5°c. per kilometre. thealtim- eter, thus graduated, is sufficiently accurate for most purposes, and other possible methods, such as the timing of a sound wave emitted from an aircraft and reflected from the ground, have not yet shown that the greatly increased complexity of the appa- ratus and its use, is justified by additional accuracy, except, perhaps, at very low altitudes. speed indicator —the speed of the aircraft through the air, at any height, is deduced from the reading of an air speed indi- cator. this instrument has a diaphragm, the movement of which is controlled by the difference of pressure in two tubes, which are set parallel to the fore and aft axis of the aircraft, and as far as possible from the effect of irregularities of airtlow due to the aircraft structure. one of the tubes registers the static pressure of the atmosphere by means of a series of small holes drilled circumferentially in the tube, whilst the other has an open end facing the direction of motion of the aircraft and registers the sum of the static pressure and the pressure due to the motion of the aircraft. the movement of the diaphragm is, therefore, due only to the kinetic effect of the movement of the aircraft relative to the air, and is a function of the product of the air density and the square of the speed through the air. since, however, the air density decreases with increase of height, the reading of the air speed indicator must be multiplied by a factor 1g depending upon the height in order to determine the true speed through the air. this is most conveniently performed by the use of a circular calculator fitted circumferentially round the dial of the air speed indicator. the aircraft compass.—the desiderata of any type of com- pass are (a) that there shall be a datum line indicating a known direction; () that a restoring force shall be set up if the direc- tion of this line is altered by some agency external to the sys- tem; and (c) that the oscillations natural to the system after a disturbance shall be damped. the magnetic compass is par- ticularly suitable for aircraft, since these requirements can be provided without prejudicing lightness and simplicity. when first adopted for aeronautical purposes, however, it proved far from satisfactory, owing to the considerable oscillation of the needle in approximately straight flight and its apparently erratic behaviour on turns from north and south. northerly turning error —the explanation of the jatter phe- nomenon, known as the northerly turning error, was given by keith lucas at the royal aircraft factory. except near the equator, a compass needle will not lie horizontally if pivoted at its centre of gravity and, in the latitude of london, a small weight must be added to the south end of the needle in order to balance the tendency of the vertical component of the earth’s magnetic field to make the north end dip. if an aeroplane com- mences to turn from a northerly course, the effect of the centrif- ugal force on the unbalanced weight is to rotate the compass card in the same sense as that in which the lubber line of the compass is rotating in azimuth. if the angular velocity of the card is greater than the rate at which the aeroplane is changing course, the compass will indicate a turn in the apposite direction to that actually occurring. the unbalanced weight may be dis- pensed with by double-pivoting the compass needle so that it is horizontal when the aeroplane is level, but the northerly turn- ing error will still be present on a banked turn, since the vertical component of the earth’s magnetic field will be unbalanced and the north end of the needle will dip unless the aeroplane is on an even keel. the error can be reduced by choosing a long period and a high degree of damping for the magnetic system; since, however, both these factors increase the time of settling down to the true reading after a displacement, there is a limit to the extent to which they can be employed as correctives. the oscillations of the compass card, when the aeroplane is executing a series of small banked turns in approximately straight flight, is also due to the presence of the unbalanced weight. if the damping coefficient is low, and the periodic time of the compass is approximately equal to the period of the aero- plane oscillation, the amplitude of the compass card oscillation may be very large; but it has been found that if the damping exceeds 60% of that corresponding to a periodicity, the oscilla- tions are small, whatever value the period of the aeroplane oscillation may have. provided that its exact reading is disre- garded during and shortly after a turn, and that it is not installed in close proximity to magnetic material, the magnetic compass is now sufficiently reliable for use in aircraft. effect of sitcel—the increasing use of steel in aeroplane struc- ture and equipment would, however, appear to render inevitable either the installation of the magnetic compass in the tail of the aircraft or its substitution for another type, though probably not for all types of aeroplane. for one particular magnetic con- dition of the aircraft a simple and sufficient deviation correction may be made (see 6.805). if, however, any moving component is constructed of steel, or if, owing to vibration or other cause, the magnetic condition of the steel parts varies during a long flight, the reading of the compass may be quite unreliable. installa- tion of the compass in the tail of the aircraft would necessitate means for communicating the reading to an instrument fitted on the dashboard. one such device is the selenium compass, which records whether the pilot is maintaining a predetermined course but cannot conveniently be used for taking bearings. earth inductor compass.—another instrument, the earth in- ductor compass, developed in america, depends upon the measure- ment of the electromotive force induced by the earth’s magnetic 20 field in a coil rotating about a vertical axis. one development of the principle utilises an armature rotated by a cup ane- mometer, the current being collected by four brushes so con- nected to a galvanometer on the dashboard that a null reading is given when the pilot is flying on a predetermined course. an advantage of this instrument consists in its having prac- tically no ‘‘ memory,” i.e, the compass behaves normally almost immediately after a turn. it weighs much more than the magnetic compass, however, and suffers from lack of sensitivity. the gyro compass.—another possible alternative to the magnetic compass is the gyro compass (see gyroscope), which is now being used for the navigation of large ships. it is improbable, however, that the type suitable for maritime navigation will ever be employed on aeroplanes. an alternative type, which may well prove suitable for aeronautical purposes, has been made by s. g. brown. in this instrument no means are employed whereby a restoring force is set up if the direction of the axle of the gyro, relative to the earth, is displaced, and its efficacy is dependent upon the extent to which disturbing forces, such as friction, can be eliminated. position finding —the utility of dead reckoning navigation is limited by the need of at least occasional observation of the ground, and, unless it is possible definitely to recognise some landmark from time to time, the method is subject to accumu- lative error. in fog, or when an airman is flying above cloud, the ground is not visible and alternative means of navigation must be provided. one such means, direction finding by wireless, (see wireless) is still the subject of experiment; while another, navigation by astronomical observation, has been ysed at sea since the 16th century. wireless direction finding.—all methods of direction finding by wireless have one underlying principle. if a rectangular coil, capable of symmetrical rotation about a vertical axis, be so placed that the plane of the coil is at right angles to the direction from which the wireless waves are transmitted, no current is set up in the coil; in any other position an oscillating current flows having a maximum value when the coil is set “ edge on,” 1.e., when the two vertical sides and the sending station are in the same straight line. if, therefore, the coil be rotated until the current is observed to be a maximum or minimum, the bearing of the sending station can be determined. in practice, telephonic detection is employed and the position of the minimum or zero signal is preferred to that of maximum strength owing to the much greater sensitivity obtainable. for ground reception the bellini-tosi system is usually employed; this consists of two small fixed coils, mutually at right angles, each being connected to an aerial loop, and a rotatable search coil symmetrically arranged with respect to the two fixed coils. ground station d.f.—this is the only method at present regularly employed on commercial air routes. the aeroplane carries a transmitting set, and, by wireless telephony, calls for its position. two or more ground stations observe the bearing of the aircraft, and one of them plots the position on a chart and communicates the information to the aeroplane. this method is fairly satisfactory for day use, but reception by a ground station at night is subject to inaccuracies known as “ night-effects.” moreover, it is for many reasons desirable that the position should be determined by the aircraft itself and not by a ground station. for this purpose it is suihcient if the ground station transmits signals and the aircraft carries a recciving set, together with means for detecting the direction of the wireless waves. the following are various modifications of this system; each has been the subject of experiment but none is yet in regular use. wing coil d.f.—this method has undergone successful air tests and can be operated by the pilot. two sets of coils mutually at right angles are fitted in the aeroplane wings and connected to a receiver. one set, the main coils, is directed fore and aft, and the other, the auxiliary coils, is athwartships. when the aircraft is headed toward the sending station the auxiliary coil receives no current but the main coil receives maximum energy. under these conditions reversal of the auxiliary coil has no effect upon the signals received by virtue of the current set up in aerial navigation the main coil. if now the aircraft be turned so that its fore and aft axis is no longer toward the station, a current is set up in the auxiliary coil, and alternate reversals of the auxiliary coil will, under these conditions, result in unequal strengths of sig- nals; the aircraft must then be turned until equality of signal strength is obtained. the disadvantage of the method is that it 1s convenient only when the sending station is the objective or lies on the desired track of the aircraft; moreover, the time taken to ily to the sending station when navigating by wing coils will not, in general, be the shortest possible time, though this is not a very serious objection. the method cannot conveniently be employed to determine position; “‘ homing” by wing coils merely ensures that the aircraft will eventually reach the send- ing station. . fuselage coil d.f.—an alternative method is to use two coils mounted on frames suitable for fitting in the fuselage of the air- craft; these coils are mutually at right angles and may be rotated about a vertical axis. with this arrangement the bearing of a sending station may be obtained, without change of course, by the method described for wing coils. two or more such bear- ings on different sending stations, of known positions, would enable the position of the aircraft to be determined. this method necessitates special apparatus and requires the attention of a skilled observer. directional transmission.—in this method the ground station apparatus consists of a fairly large loop which rotates at a speed of about one revolution per minute. no energy is radi- ated in a direction at right angles to the plane of the coil, and by “tuning in” to the station and observing the time between two consecutive points of silence, the exact speed of rotation can be estimated. further, if the station emits a dis- tinctive signal when the loop is directed, say, due north, the bearing of the sending station can be calculated by observing the time interval between the distinctive signal and the next zero signal. in practice it would be necessary to use two dis- tinctive signals, preferably 90 degrees apart. when fully developed, position finding on the aircraft itself will probably be more satisfactory than the existing system of ground reception, particularly for night use. it seems fairly well established that errors due to “‘ night-effects ’’ are confined chielly to ground reception and are quite small when the aero- plane itself is used for reception. navigation by astronomical observation.—the determination of the position of an aeroplane by astronomical observation is attended by many difficulties and is hardly likely to be employed on a developed air route along which it is convenient to erect wireless beacons. tor long distance pioneer flights over unde- veloped country, or, for example, for trans-atlantic flights, astronomical observation remains the only method of checking the d.r. position. the principles employed are essentially those used for maritime navigation (see navigation), but both the accuracy needed and the accuracy attainable are considerably less on aeroplanes. determination of the altitude of a heavenly body necessitates either that the horizon shall be visible, or that the true horizontal or vertical be otherwise known. the horizon is often indefinite from an aircraft, and this fact has led to the development at the royal aircraft establishment, farnborough, of an ingenious instrument known as the bubble sextant. the r.a.e. bubble sexiant—the image of a bubble, con- tained in a chamber having a spherical glass cover, is observed by reflection from a pentagonal prism and subsequent trans- mission through a collimating lens and an index glass. the optical distance of the bubble from the lens and the radius of curvature of the glass cover are both chosen to be equal to the focal length of the lens. this arrangement provides that if an image of the sun, obtained by reflection from the index mirror, is brought into coincidence with the bubble image and the instrument is rocked in the hand, the bubble remains in focus, and bubble and sun appear to move together. when measuring the altitude of a star it is more convenient to observe the bubble image by reflec- tion from the under-side of the index mirror and the star by aerial navigation direct transmission through the mirror. the bubble is, of course, subject to displacement if the aircraft experiences an accelera- tion which is not truly vertical and, in practice, an average of ro readings is taken. the degree of accuracy essential for maritime navigation is greater than that required for air navigation and it has been found convenient to substitute reduction of observa- tions by logarithmic tables for more rapid methods, the most important being a compact cylindrical slide rule, devised by l. c. bygrave, which is capable of giving an accuracy of at least two minutes of arc. for airship navigation the conditions approxi- mate more to those appertaining to maritime navigation and there is no reason to suppose that navigation by astronomical observation will not generally be employed on airship routes. navigation in fogs and at night flying in fog.—the safety of an aeroplane flying near the ground depends largely upon the pilot’s ability to maintain, within certain limits, an even keel. moreover, for navigational purposes it is desirable that a straight course should be main- tained. under normal conditions this presents little difficulty, but in fog, when visual connection with terrestrial and celestial objects is not possible, means should be provided to enable a pilot instantly to recognise any involuntary change of course. it is not at first obvious why the natural gravitational force is not a sufficient indication of the vertical. imagine an acroplane to be at rest on the ground, with its wings horizontal, and an ordinary spirit level to be mounted laterally, so that the bubble is in the centre of the tube. if now the aircraft be tilted so that one wing is higher than the other, the displacement of the bubble from the centre of the tube will indicate the nature of the tilt of fic. 1.—principle of the turn indicator. the aircraft. change of course of aircraft in flight is normally associated with a lateral tilting of the wings, and, if a bubble level indicated the tilt, or bank, as it does on the ground, depar- ture from straight flight would be shown by a displacement of the bubble. when an aircraft is turning, however, it is flying on a curved path and the centrifugal acceleration, which acts at right angles to the direction of flight, needs to be considered. it is naturally not possible for a bubble level to distinguish acceleration due to one cause from that due to any other, and the liquid in the level is acted upon, not by the earth’s gravitational force alone, but by the resultant of the centrifugal force and the earth’s gravitational force. this resultant acts at right angles to the ‘“‘ floor” of the aircraft and the bubble consequently seeks, not the true vertical, defined by gravity, but the apparent vertical, eo | defined by the direction of the resultant of the accelerations. if, therefore, the aircraft executes a banked turn in flight, the bubble remains in the centre of the level and does not indicate a change of course. in fog or cloucdl a pilot may be unaware that his craft is turning and the magnetic compass affords him little o 6 aileron =< control indicator spat.ng so38s/18 miles noua 0°24 @ fic. 2.—the “' reid ’’ turn indicator. | aid, since, as already explained, it is unreliable under these con- ditions. the need for an instrument which will indicate that the aircraft is changing course has led to the development of the gyro turn indicator. the gyro turn indicatoy.—the gyro turn indicator provides a pilot with a continuous indication either of the maintenance of straight flight or, if the aircraft 3s turning, of the direction and approximate rate of the change of course. the action of the instrument may be understood by referring to fig. 1. a gyro wheel, a, rotates about an axis, b c, which 1s horizontal in level flight, and is usually arranged to be athwartships on the air- craft. the frame carrying the wheel and its axle is free to rotate about another horizontal axle, de, at right angles to bc (i.e., in the fore and aft line of the aircraft). the axle, de, supports the wheel and the inner frame in an outer frame which is fixed to the aircraft. if the aircraft changes course, the whole system 1s rotated about a vertical axis, f f, and, in accordance with the laws governing the behaviour of a gyroscopic system (sce 12.769) the wheel and the axle, b c, commence to rotate about d e. this rotation, which is known as precession, is controlled by springs, g g, so that the deflection of b cis a measure of the rate of change of course and so that the axle, bc, may return to a hori- zontal position when the aircraft is again flying on a straight course. the deilection of b c may be indicated to the pilot by any suitable means. the instrument has proved of the greatest — use for flying in fog or cloud; it has been developed in a number of countries, but all types depend upon the principle outlined above and differ chiefly in the methods of indicating the turn and driving the gyro wheel. one type, the reid turn indicator, is shown in fig 2. this instrument employs two rows of electric lamps, each row consisting of four red and four green lamps on each side of a central white lamp. ‘the upper row is controlled by the position of mercury in a u-tube and indicates whether the aircraft is correctly banked. the lower row is controlled by the gyro and indicates the rate of turn. to maintain straight flight the pilot must fly so that no coloured lights are visible. the gyro rudder conirol—another device, making use of the gyrostatic principle, is the gyro rudder control developed at the royal aircraft establishment, farnborough. the primary object of this instrument is to relieve a pilot from the strain 22 consequent upon the maintenance of a straight course for a long period of time. the principle is shown in fig. 3. the gyroscopic system consists of a rotor, a, mounted in an inner gymbal ring, b, which is pivoted about a horizontal axis to the outer gymbal ompressed air thiet pipe gyrobax rotatably mounted on pivot fic. 3.—r.a.e. gyro rudder control. ring, c. the latter carries a pin, d, and rotates about a vertical axis, thus opcrating a valve bv means of the lever, e. the system is mounted on a vertical pivot, f, and two ratio arms, one attached to the false rudder bar and one to the gyro box, engage at g. if the aircraft turns, for example, to starboard, air is admitted to the port cylinder, shown in fig. 3, and the starboard cylinder is put into communication with the atmosphere. the pistons consequently move and operate the false rudder bar pivoted at h, which in turn operates the pilot’s rudder. movement of the false rudder bar, however, causes a rotation of the gyro box on its pivot, f; the valve then closes and the rudder bar returns to its neutral position. this instrument has proved very satisfactory during air tests and is likely to reduce considerably the strain of piloting an aircraft over a long distance. landing in fog.—the problem of effecting an aeroplane land- ing in fog is still the subject of research and experiment. apart from the necessity of navigating and controlling the air- craft, connection with the ground must be established in such a manner that a landing can be effected. one promising method, which, however, has not yet reached the stage of full scale experi- ment, is the use of a cable, laid just beneath the ground, excited by alternating current. this cable forms a closed circuit on the aerodrome, and a convenient size and shape would be two parallel sides of about ? m., joined at the ends by two semi-circles of ¢-m. radius. the magnetic field above the cable will consist of circular magnetic lines of intensity varying approximately inversely as the height above the cable. vertically over the cable the lines will be horizontal, and if instruments are carried enabling this condition to be detected, the pilot will be able con- tinuously to fly vertically over the cable and by gradually de- creasing his height to effect a landing. the instruments carried on the aircraft may consist of two search coils oriented in such a manner that when the aircraft is vertically over the cable the resultant electromotive force induced in them is zero, but if the aircraft departs from the de- sired course, the e.m.f. induced in one coil will be greater than that induced in the other and the instrument will indicate this condition. the height can be determined by a suitable cali- bration of an instrument which measures the intensity of the magnetic force. here is no reason to suppose that this method, when fully developed, will not be largely adopted for the land- ing of aeroplanes in fog, but the system is, of course, limited to prepared aerodromes. another promising method is the employ- ment of ‘lanes’ of neon lamps laid on the aerodrome; under - conditions of low visibility neon lamps have becn found, by experiment, to have a greater visibility range than any other form of illumination. aerodrome flying and landing at night—navigation at night presents far less ditliculty than do flying and landing in fog. there will usually be no difficulty in navigating by d.r. methods, and the problem of night landing is mainly one of supplying suflicient and suitably placed illumination at aerodromes. it is probable that a night-operated air route will have a number of light beacons at suitable distances apart and, unless visibility is very poor, the pilot’s task will be merely to fly from beacon to beacon. the united states has a lighted airway of 1,900 m. from new york to rock springs, consisting of a number of terminal aero- dromes about 250 m. apart, each having two electric arc beacons of 500,000,000 candle-power each, one circling the horizon every 20 seconds and the other providing tlood lighting for landing purposes. in addition there are emergency fields about 15- 30 m. apart, each provided with a light beacon; acetylene gas beacons have been installed every three miles over a consider- able proportion of the route. a thousand million candle-power beacon near dijon, france, having a normal range of about 300 m. has been built mainly to assist aircraft flying at night between paris and algiers. an interesting method of facilitating night landing, known as the hoenig circles, originated in germany. the installation consists of two vertical lighted circles of different sizes, placed one behind the other, mounted on a trolley and so arranged that the common axis of the circles indicates the direction of janding. the pilot manoeuvres his craft until he observes two concentric circles; by preserving this orientation of the circles a correct landing can be effected. navigation of atrships—whilst the principles underlying the navigation of aeroplanes apply also to the navigation of air- ships, the foregoing description of navigational methods and of the peculiar advantages and disadvantages associated there- with, applies chicily to heavier-than-air craft. an airship is much steadier than an aeroplane, and instru- ments, particularly the magnetic compass, are more reliable. under suitable conditions, astronomical observation can be taken and reduced without difficulty and the accommodation for wireless equipment is greater than in an aeroplane. it will be seen that the conditions obtainable for airship navigation are substantially those employed in maritime navigation. it is important to note, however, that the navigation of airships over long distances will render the availability of accurate meteorological information of primary importance, since an air- ship, unlike an aeroplane, can land only at prepared aerodomes. bibliography.—~text-books on aerial navigation: s. f, card, air navigation notes and examples (1919); j. e. dumbleton, aerial natigation (1920); h. e. wimperts, primer of air navigation (1920). the last-named includes descriptions and illustrations of the aero bearing plate, the wind gauge bearing plate and the course setting sight. wireless direction finding: niermann, funkentelegraphte fiir fliugzeuge (1921); r. keen, direction and position finding by wire- less (1922). compasses: t. w. chalmers, the gyroscopic compass (1920); p.r. heyl and l. j. briggs, the earth inductor compass (american phil. soc., april 1922); the selenium compass is described in la nature of sept. 2 1922 and in the aerial age weekly of aug. 29 1921. position finding by astronomical observations: astronontical afeth- ods in aertal navigation (national advisory committee for aero- nautics), report no. 198 (u.s.a.); comm. t. y. baker and maj. l. n. g. filon, “ position fixing in aircraft during long distance flights over the sea,”’ trans. roy. aero. soc. (1920), gyro rudder control: dict. of applied physics, vol. 4, p. 67. landing by hoenig circles: zeitschrift fiir flugtechnik (sept. 1925). descrip- tions of long distance flights: comm. g. c. westervelt, comm. h. c, richardson and lt.-comm. a. c. read, the triumph of the n. c.’s (1920); air comm. e. m. maitland, the leg of hf, m. a. r34 (1920); sir ross smith, 14,000 miles through the atr (1922); sir a. whitten brown, “ flying the atlantic in sixteen hours,” the royal atr force and civil aviation record, vol. 2 (1921). (a. p. r.)} aerodrome (gr. aer, air; dromos, a course).—at the con- clusion of the war 1914-8 the international commission for air navigation, which was a sub-commission of the peace conference and is now a part of the organisation of the league of nations, was appointed, and drafted regulations, which are known as “the convention relating to international air navigation.” aerodrome provisions of the convention by virtue of this convention, each contracting state undertakes, in time of peace, to provide aerodromes equipped with ground organisation, to include cus- toms posts where required for the operation of international air transport. every aircraft engaged in international transport must, before proceeding to a foreign country, be cleared for customs, and therefore depart either from a customs acrodrome in the first place or land at a customs aerodrome before leaving the country; similarly, aircraft arriving from abroad must make their first landing at a customs aerodrome. although it is not definitely stated in the regulations that hangar accommodation, etc., should be provided, the custom has been universally adopted that each state shall erect, at its own expense, aerodromes fully equipped in every respect. details as to the greater part of this equipment are laid down in the regulations so as to obtain as nearly as possible international uniformity. national regulations—the majority of sontmetine states have issued their own regulations. in the british regulations, which are laid down in the air navigation order, 1922, and the air navigation (consolidation) order, 1923, an aerodrome is defined as follows:— an acrodrome means any definite and limited ground or water area intended to be used either wholly or in part for the landings and departures of aircraft. aircraft in this connection include all balloons, whether fixed or free, kites, airships or flying machines. all aerodromes in great britain and northern ireland, which are used as regular places of landing or departure by aircraft carrying passengers for hire or reward, must be licensed for the purpose by the secretary of state for air, and, further, every aerodrome which is licensed for public use, or which is open to such use by british aircraft upon payment of charges, shall to the same extent and on the same conditions be open to use by aircraft possessing the nationality of a contracting state to the convention for the regulation of aerial navigation. other countries have regulations with regard to the acrodromes which exist in their territory. definition.—although the origin of the word sepodeome only refers to the area to be used for the purposes of alighting and taking-off and is supported in this by the definitions in the regulations, in general practice the word aerodrome is used to include, not only the ground or water, but also the buildings and installations required for the proper operation of air traffic. provision of acrodromes.—on a recognised air route for the operation of land machines, such, for example, as that from london to the continent, a series of aerodromes is provided at an average distance apart of 25 m. and these are classified, ac- cording to the purpose which they serve, as terminal aerodromes, aerodromes, landing grounds and emergency landing grounds. these are situated at croydon (terminal aerodrome), penshurst (landing ground), marden (emergency landing ground), lympne (aerodrome) and littlestone (emergency landing ground). lympne and littlestone are approximately 5 m. apart and the reason for the existence of the latter, which is sited about sea-level near the coast, is that the former, being on high ground formed by the cliffs overlooking hythe, is liable to be obscured by low clouds which drift in from the sea. it is essential that air- craft flying over the channel shall have a landing ground near the coast for use in case of emergency; hence the provision of al- ternate grounds, one at high and the other at low level, both of which are unlikely to be obscured by cloud or fog at the same time. of the above, both croydon and lympne are provided with customs facilities and in consequence are further classified as *“ customs aerodromes.”’ equipment.—a terminal aerodrome normally 1s equipped with hangars for the housing of aircraft, workshops, bonded and technical stores, administrative and booking-offices for the companies operating services from the stations, as well as for travel and transport agencies, petrol and oil companies, etc. in addition facilities normally exist for dealing with customs and for carrying out inspection by the immigration authorities. such aerodromes are further provided with wireless and mete- 23 orological facilities, together with government personnel to operate them. various types of illuminating beacons for the assistance of aircraft flying by night, in addition to some form of device for indicating the direction of the wind both by day and by night, are installed. accommodation is provided for an official belong- ing to the aeronautical inspection dept., whose duty it is to ensure that aircraft are being properly maintained, for traffic hands who carry out porterage duties and assist in handling machines on the ground, and for the officials appointed by the govt. for controlling air traffic, the aerodrome and the govern- ment personnel employed there. landing grounds——landing grounds may, or may not, be equipped with hangar accommodation, but are normally pro- vided with beacons for indicating their position by night and wind indicators to show the direction of the wind both by day and night. telephones are generally installed, but no personnel other than a caretaker will be found there. emergency landing grounds are rarely equipped with any form of hangar accommoda- tion and may, or may not, have any form of installation for in- dicating their position by night or the direction of the wind. each of the above-named stations, with the exception of marden, has its name in large capital letters marked on the surface of the aerodrome in chalk and the good landing area is outlined with automatic red flashing lamps which operate day and night. sea traffic aerodromes.—aerodromes for the use of sea-going aircraft, such as flying boats or seaplanes, differ from those al- ready alluded to mainly in having an open-water site from which to depart and on which to alight, and slipways leading from this water site to the hangar accommodation and to the prepared passenger departure and arrival platforms. for aircraft of this nature emergency landing grounds are unnecessary since the whole of a water route is itself an emergency landing ground. alrship :lerodromes.—aerodromes for the use of airships will in general be provided with all the facilities outlined above as requisite for the operation of land aircraft, but in addition will be equipped with a mast at which an airship can be moored, known as a “ mooring mast,” and a gas plant for the production of the gas required for inflating the ship. under normal con- ditions an airship will only be berthed in a hangar for purposes of overhaul or repair in the same way as a ship goes into dry dock; at other times it will invariably be moored at a mast. the modern mooring mast consists of a steel girder structure rising to the height of approximately 195 feet. the head of the mst is fitted with a receiving arm into which a cone on the bow of the airship engages; this allows free movement in any direc- tion. the mast is supplied with lifts for the use of the crew, passengers, goods, etc., and with pipes for the supply of fuel and water. a passenger mounting in this lift will, on emerging from it, find himself in a small compartment from which a covered way leads into the nose of the airship and thence down the passenger accommodation. this covered way may be likened to the connection between two railway coaches on a corridor train. aerodrome condilions—leaving out of consideration emer- gency landing grounds, which require as a minimum two landing runs at right angles to each other of not less than goo yd. long by 200 yd. wide, an aerodrome for the use of aeroplanes should normally consist of an area not less than 800 by 800 yards. the surroundings should be as free as possible from all obstruc- tions, such as high buildings, trees, chimneys, wireless masts, etc. the surface should be level and the soil not liable to become boggy or marshy after heavy rain. in europe aerodromes are invariably grass-covered; in tropical climates, however, they frequently consist of hard sandy surfaces or bare earth. in the choice of a site for an acrodrome several factors must be considered, but as it is rarely found possible to combine all the desirable qualities in one place the final selection is generally largely a matter of compromise. accessibility to the centre which an aerodrome serves is of primary importance, but in the neighbourhood of a populous area such as, for example, london, a large open space of the size required can normally only be ob- tained on the outskirts of the city. under these circumstances, it is highly desirable that rapid and easy means of communication 24 shall exist connecting the aerodrome with the city. a considera- tion of paramount importance, in the present stage of develop- ment of air transport, is that an aerodrome designed for its use should be sited in such a position as to be as free as possible from fog, town smoke or local mist. for preference the country surrounding an aerodrome should be as open as possible. two examples of main terminal aerodromes may be quoted: london terminal aerodrome—the london terminal aero- drome which is situated at croydon, a distance of approximately 13 m. from hyde park corner. the distance to the centre of london is far too great, there is no direct thoroughfare which leads from the centre of london to the aerodrome, and it is not fed with direct rail or bus communication. on the other hand it is freer from fog and mist conditions than almost any other locality in the immediate neighbourhood of london. before croydon was finally decided upon, exhaustive investigation was made into many other sites. an aerodrome in the north of london, unless having overwhelming attractions, was undesirable owing to the fact that aircraft proceeding overseas would find it necessary, either to fly over london or else make a circuit of it, in elther case adding appreciably to the distance of the journey and rendering machines liable in foggy weather to be unable to reach their destination, owing to having to traverse the smoky and prevalent fog area extending east and west following the general line of the run of the thames. in other districts in- vestigated, in some cases it was found to be impossible to obtain sufliciently large areas, in others fog was prevalent, while in others the nature of the ground did not permit of their being satisfactorily converted into an aerodrome. berlin terminal aerodrome.—the terminal aerodrome at berlin is situated at tempelhoferfeld and is probably one of the most perfectly sited aerodromes in the world. this city had the advantage of possessing an enormous parade and review ground which is on high ground, is free from fogs, is almost perfectly flat, is fairly clear of obstructions surrounding it, is on a main road, along which trams pass, and is within ten minutes drive of the heart of the city, a perfect siting for the main terminal acrodrome of the country. (i. a. e. e.) aerodynamics.—the word aerodynamics refers to those branches of dynamics which form the foundation of the study of aeronautics; it thus applies to two distinct branches of this subject, which relate respectively to the reactions between air and solid bodies moving through it and to the movements of bodics under those reactions and the force of gravity. the former is a development of a mathematical study which has gen- erally been known as “ i[ydrodynamics,”’ because, although it applies equally to all fluids, it was developed originally with the object of throwing light upon the movements of water; the latter is a branch of “ rigid dynamics,” the study of the movements of rigid bodies of appreciable size under the action of external forces. in no single instance has the reaction between a fluid such as air and a body moving through it been deduced directly from the laws of motion. the equations of motion can be formulated, but they have not yet been solved precisely. aeronautics therefore relies primarily upon direct measurement of these reactions, but unfortunately their precise measurement upon the full scale requires, in general, experiments in free flight, which are difficult, costly, and sometimes hazardous to carry out. for these reasons nearly all the detailed data available for the study of aeronautics have been obtained from small models of the parts concerned, suspended in artificially induced currents of air, which may either flow down large pipes, called wind tun- nels, or take the form of a spout of air directed through the experi- mental chamber. the technique of the measurement of reactions upon models in a current of air has reached a high development in many countries, and the world’s output of data relating to such experiments is now very large. the main function of the theoreti- cal study of fluid dynamics in its relation to aeronautics has been therefore to indicate the extent to which such experimental data can be applied with confidence to the full scale, and to extend the range of phenomena over which a single experiment can be made to give useful information. aerodynamics principles of dynamical similarity—these considerations lead at once to the study of similarity in fluid motion. to what extent will the flow round a large body moving fast through air possibly of low density, such as is found at great heights, be similar to that about a small model of that body moving at a slower speed through air of greater density, such as is found near the ground? experiment has led to the conclusion that the mo- tion of a fluid past a body of given form, moving through it at a given speed, is dependent only upon three physical properties of the fluid: density, viscosity and compressibility. no other property of the fluid is known to have any appreciable influence upon the motions with which aeronautics is concerned, nor does it appear to matter whether the fluid is a liquid or a gas. these perhaps surprising conclusions have results of immense impor- tance, for from them can be deduced the laws governing the application of model data to full scale problems, and it becomes possible to carry out model experiments in any fluid which may be most convenient; moreover, the accumulated knowledge of “‘hydrodynamics”’ becomes directly applicable to aeronautics. the precise statement of the laws of similarity, which can be deduced from these premises, 1s as follows:— let 7 be some length defining the size of the body, v the velocity of the body through the fluid, and p, zand e, the density, viscosity and bulk elasticity of the fluid. then strictly similar motions can occur round two bodies of the same shape moving similarly, only when the two quantities p and ae are the same for both motions.‘ quantities of this nature, which must not change if dynamical similarity is to be possible, are called “criteria of similarity ’’; they are always non-dimensional, and hence have the same numerical value, no matter in what system of units they are expressed, provided that the system is dynamically consistent.” a the criterion <3 constant, expresses the fact that in similar motions the volume strains in the fluid at similarly situated points must be the same. it may be re-written ¥ constant, where a is the velocity of sound in the fluid, for @? is proportional to th general, such strains are small, unless the velocity of some of the bodies concerned rises above about =, which in air is about 550 ft. per sec.; it is therefore usual in practical aeronautics to treat air as incompressible, unless velocities exceeding about 500 ft. per sec. are present. such velocities have hitherto occurred in aeronautics only near the tips of high speed air-screws, hence it is in this connec- e a ad v . 9 * ° tion only that the criterion, — constant, has had any practical signifi- cance.2 when this criterion is of importance it leads to the conclusion that model experiments in air should be carried out at the same speed as the full scale motion represented, but that, so far as compressibil- ity is concerned, the size of the model is immaterial. the magnitude of the criterion p-—called the ‘‘ reynolds number ”—determines the part played by viscosity in the motion. if two motions about similar bodies have the same rey- nolds number the effect of viscosity in the two motions will be the same. if the reynolds numbers are not the same, viscosity will play a more important part in that motion for which the number is the smaller. thus the effect upon a motion of increas- ing either the density, the velocity or the scale, is the same as that of decreasing the viscosity. a fluid which appears rela- tively viscous to the slow passage of small bodies, may therefore appear to be very inviscid to the rapid passage of large bodies. air, from the point of view of a gnat, is a somewhat viscous fluid; to an aeroplane it is much more mobile, whilst in relation to the general movements of the winds over the earth it is extremely inviscid. the aeronautical engineer may be more interested in the forces generated by air motions than in the form of the motions them- selves, but when a series of motions are known to be similar, the 1 these conclusions can be reached solely from newton’s laws of motion and from considerations of the meaning of similarity. 2a dynamically consistent system of units is one in which new- ton’s law—force proportional to mass acceleration—can be ex- pressed as an equality. 3 for experiment on air-screws with tips moving above the spcec! of sound see reports aeronautical research committee, no. 884, 1403 4, vol. i, p. 345. aerodynamics laws governing the variations of force from motion to motion are easily deduced from the fact that all quantities, or combinations of quantities, having the same dimensions, vary in the same way. forces upon similar areas will thus vary in proportion to p77/*, or to uv/, or to p’p, because all these combinations of the inde- pendent variables have the dimensions of force. if, as in most aeronautical problems, the reynolds number p~- is the only cri- terion of similarity which need be considered, and if it has the same value for a series of motions of fluids past bodies of the same shape, then these motions will be strictly similar, and the forces acting upon similar areas will vary as above. i{, however, the criterion varies from motion to motion, strict similarity is impossible, because viscosity will playa moreimportant part in one motion than in the other; nothing can then be ascertained as to the mode of variation of the forces, except by detailed calculation or experiment. , this condition can be expressed! by the equation f = pv7/7f [ | where f is the force under consideration and sf os stands for any function the form of which can be determined cither by detailed calculation or by experiment upon the effect of altering any one of the four variables concerned. ‘the effects of a change in any one of the four independent variables can thus be ascertained by experiment upon the effects of varying any other of the variables,? always pro- vidled that there is, as is generally true in acronautics, no other criterion of similarity than the reynolds number to be taken into account. for such values of p, v, 7 and u, as occur in practical work with ® e pp t * e a,e ic] man-carrying aircraft, f| —— | is very insensitive, even to consider- able changes in the value of the reynolds number, and calculations relating to free flight are generally made on the assumption that this function may be treated as a constant, so that we have f =kpv°p where k is dependent only on the shape and mode of motion of the body through the air. when, however, deductions have to be made from small scale model experiments, this assumption is not always sufficiently accurate. with values of / and v which are practicable in wind tunnels, small though appreciable variation in & with change of reynolds number are the rule rather than the exception, and the experimental data are then said to be subject to “ scale effects.” the reynolds number of a motion on the full scale is generally very much larger than that of the corresponding model experiment, even in the largest and most powerful wind tunnels available,’ and some uncertainty in the prediction of the behaviour of aircraft from wind tunnels data thus occurs; for there is, as we have seen, no known method of predicting the changes in k consequent upon a change of the reynolds number. a long series of carefully selected experiments, carried out in free flight upon the full scale, has shown that much of the data of importance in acronautics is not subject to serious ‘ scale effects ” between the highest reynolds numbers conveniently obtained in ordinary wind tunnels and the full scale; those types of data which are subject to serious scale effects are becoming more and more clearly defined as research proceeds. the coefiicient # above, is called a non-dimensional coefficient* of force, and the force to which it refers is indicated by a sufiix. thus ky = f - j ev-p 1 this condition might equally well have beet expressed by either non-dimensional coefficients of other quantities, such : vi 2 vi | of the equations f =2vio( 0“) orf = hao where ¢and wrepre- sent other functions; each of these forms is useful in certain cir- cumstances, but that given in the text is generally used in aeronau- tics, for a reason which appears below. i 2for example, should it be found that f«v" then f | = vi n—1 n jv bs = and pee ub in america a wind tunnel has been made which works in com- pressed air for which * is many times greater than in air at atmos- pheric pressure. this allows a reynolds number equal to that on the tull scale to be obtained in many instances. at the time of writing the accuracy with which this tunnel faithfully represents full scale conditions in free flight in still air, is under discussion. * also called an ‘ absolute coetficient.”’ 5 the symbol ? may be replaced by any other quantity which defines the size of the body and is of the dimensions of area, thus f . . a . s aves" where s is the wing area, is used in relation to aeroplanes and f pv?a ; » where a is the total volume, is used in relation to airships, a5 as pressure per unit area (p), and moment about an axis (m) are formed in a stnular manner, thus:— ie ee ee ap = t5357 : = hi pv? mn pv3 all such coefficients are constant for strictly similar motions, and being non-dimensional their value is not affected by changing from one consistent system of units to another. they are therefore much used in theoretical calculation, and for recording the results of experi- ments, air reactions in relation to the theory of inviscid flnids.— let us consider the motions which occur when bodies of various shapes move through otherwise undisturbed air. bodies not specially shaped for low resistance generally drag behind them a quantity of air which, relative to themselves, is nearly motion- less. this is called “ dead air”; the boundaries of the region which it occupies are well defined near the body, but become ill- defined at some distance behind where the dead air gradually merges into a turbulent wake of irregular eddies. the pressure in this “‘ dead ” region is always lower than that of the surround- ing atmosphere. such bodies experience a high resistance to motion and are therefore avoided in the design of aircraft. certain shapes called “ streamline ” shapes, do not drag dead air behind them, and experience much less resistance to motion than those which do so; these shapes are therefore much used in aeronautics. the fish, the modern airship hull and the wing are typical examples of streamline shapes. the air-flow around moving bodies cannot yet be calculated from first principles, hence our precise knowledge of the air reactions upon these streamline shapes is drawn from direct experiment; but much light has been thrown upon their action by the theoretical study of the movement of a hypothetical fluid having absolutely no viscosity. the mathematics of this hypothetical fluid, unlike that of the real slightly viscous fluid air, is in many instances within the powers of modern mathematicians; and had been brought to a high state of development before aeronautics began to attract attention. when a streamline body has been moving steadily through air for some time, the motion of the air relative to it becomes steady and substantially obeys the laws appropriate to inviscid fluids, except in regions containing air which has passed very near to the surface of the body. very near this surface the motion is powerfully affected by viscosity, the velocity falling off as the surface is approached until, according to best authorities,® it reaches zero at 1ts surface. the layer of fluid appreciably influ- enced by viscosity appears to decrease in thickness with de- crease of viscosity until, in the conditions of interest in aeronau- tics, it becomes very thin indeed. the viscous layer is responsible for the tangential drag or “‘ skin friction ’’ upon the surface of the body, which constitutes the major part of the resistance of the best streamline shapes, and is, in some as yet unknown way, associated with the development of dead air behind bodies which are not of good streamline shape. the study of the dy- namics of this region influenced by viscosity is proceeding vigorously in most of the aerodynamic laboratories of the world. returning to a consideration of the region in which viscous forces are negligible, which in most aeronautical problems in- cludes by far the greater part of the whole region in which the fluid is appreciably disturbed by the body, bernoulli’s well-known theorem, relating to the steady motion of an inviscid incompres- sible fluid, indicates that tf the motion is such that the velocity at all points far from the body is uniform, then the pressure (p) and the velocity relative to the body (gq) at any point in the fluid, are connected by the relation £+3pq?=constant. if p and v be the uniform pressure and velocity far from the body, this leads to p-—p=3pv ‘| — (yy ; from which the pressure at any point can be deduced from the velocity, or vice versa. the maximum pressure rise is seen to be 3p v2 and to occur when g is zero. 6 report of aeronautical research committee no. 720 (stanton) proceedings of the first international congress for applied me- chanics (delf). (von karwan and burgers), p. 98, et. seq. 26 there are certain definite ways only in which an absolutely inviscid fluid can flow past a given body, with the restriction that the motion is to be uniform at all points far from the body. one of these flows only can be generated from rest; the others, if generated by some agency not contemplated in the theory and then left to themselves, will continue indefinitely without change. the flow which alone can be generated from rest has the prop- erty that, whatever the shape of the body, the pressures gener- ated upon its surface exactly balance, so that no resultant force is exerted. certain streamline shapes—for example, the airship hull moving axially and the wing moving in its attitude of no lift'—develop close approximations to these flows; hence the pressures upon their surfaces, being in close agreement with theory, approximately balance. the resistance to motion is then very low, being due merely to skin friction and to a very slight imperfection of balance of the pressures. when the attitude of a streamline bedy in a current of air changes from that which gives rise to the unique flow which alone can be generated from rest, the flow changes progressively to a new form, which may be a close approximation to one of those flows which are appropriate to an inviscid fluid, but could not be generated in it. the precise analysis of these ilows in the hypothetical inviscid fluid, is particularly tractable to modern mathematics when the motion is two-dimensional,? and a general analysis then leads to the conclusion that the resultant force upon the body is exactly perpendicular to the direction of mo- tion of the undisturbed fluid, and is larger the further the flow departs from that one which can alone be generated from rest. in the corresponding real two-dimensional flow the reaction 1s xearly perpendicular to the direction of motion, but has a small. component opposing the motion. if the magnitude of this re- action can be measured, it is easy to identify the particular the- oretical flow to which the real flow approximates; the details of this theoretical flow can then be calculated, and are generally found to agree well with the results of detailed experiment. in the absence of some such measurement of force, no precise method has yet been devised for deciding to which of the possible theoretical flows the real flow will approximate, but joukowski, observing that for a wing of conventional shape there is only one of the theoretical flows which will not give rise to very large veloci- ties at the trailing edge, advanced the hypothesis that this particu- lar flow will occur. applied with discrimination, this hypothesis gives results closely in accordance with experience. joukowski has, in addition, developed very beautiful methods for obtaining wing sections from mathematical formulae which lend themselves well to analysis of the type under discussion, and since wings with these sections are found by experiment to give results at least as good as any that have been obtained from arbitrarily de- signed wings, it is becoming common practice to use joukowski profiles, as they may be called, in actual design. a feature of many of these profiles is that their centre line forms a segment of a circle; if - be the camber of this segment, c its chord, a the incidence4 of the chord and ?-c the maximum thickness of the profile, joukow- ski's analysis leads to the following remarkably simple approxima- tions to the lift per unit length (l) of the cylindrical wing.® l/pv2c = (1+ -8t)a@ sin (a +27) m /pv2ce? =4ry where m is the moment exerted by the lift about a point on the chord distant ¢/4 from the leading edge, pis the density of the air and v the velocity relative to undisturbed air. these formulae agrce fairly well with the results of experiment upon two-dimensional wings, but are liable to give the lift slightly too high; a small com- ponent force opposing motion, called ‘‘ profile drag,” has also to be added. this profile drag has, as yet, to be determined expcerimental- ly; it may in favourable circumstances be less than # of the lift. the two-dimensional acrofoil in free space must be supposed to extend to infinity in two directions perpendicular to the direction of ag ee 1 lift is defined as the component of the air reaction perpendicular to the relative motion of the body and the undisturbed fluid. 2 by two-dimensional motion is meant motion in which all veloci- ties are parallel to a given plane and in which the flows in all planes parallel to the given plane are the same. for such a motion to be pee the solid body must be in the form of a cylinder, which may, owever, have any desired cross section. $y-¢c=maximum distance of segment from chord. 4 incidence is the angle made by the chord with the direction of the undisturbed air stream. 5 ‘these formulae are first approximations, considering ¢ and y as small quantities. alrody namics motion, but the span of a real wing is not infinite, hence the flow about it cannot be strictly two-dimensional. no exact method of calculating the flow around a finite wing has yet been devised, but lanchester suggested that, provided the aspect ratio® is large, as in real wings, the effect of changing from an infinite to a finite aspect ratio might be represented by supposing the finite aerofoil ta be working in a down current superimposed upon the motion appro- priate to two-dimensional conditions. prandtl has developed this idea into an approximate quantitative theorem, and has shown that if the distribution of lift over the span of the finite wing is elliptic, this superimposed down current will, to a first approximation, be uniform across the span and will be equal to 2ky. a is the aspect ratio and the average lift coefficient over the whole wing.” he arrives at the conclusion that the finite wing will generate the same lift as the two-dimensional wing when its incidence relative to the s s cy . 2 a undisturbed air stream is increased by a radians, and that the resistance will be increased by an amount appropriate to a rotation of the resultant force through this angle. the distribution of lift over the span of a real wing, which is generally nearly rectangular in plan form, is not exactly elliptical, but is approximately so, and in practice the prandtl formulae for conversion from a two-dimensional wing to a finite wing, namely :— kl nies ee ad = etdoes kl? x a give results which are sufficiently accurate for most practical pur- poses. by further extensions of these methods, estimates can be made of the mutual interference of a number of wings—as in a biplane or triplane—of the effect of placing the wings in a wind chan- nel or near a plane surface, and of the mutual interference of the various blades of an airscrew. by combining the hypotheses of joukowski and prandtl, it is possible to make a fair estimate of the performance of most practicable wings or airscrews, whilst subject to many types of interference from external bodies, the only special experiment required being that to determine the “ profile drag ”’ of the sec- tion or sections from which the wings or airscrew are built up. somany approximations and hypotheses are, however, involved in these calculations, that they can only be applied safely in circum- stances for which direct experiment has shown them to be appli- cable. if, for example, the wing incidence is too great, or the wing is not of a good streamline shape, ‘‘ dead air ’’ may form, and the motions cease to approximate even roughly to any flow which has yet yielded to analysis. the wing is then said to be * stalled,’ and on account of its high resistance becomes an un- satisfactory organ of flight. little or nothing is known about the causes which make a wing stall and the stalling angle has still to be determined by direct experiment. various methods of delay- ing the stall have, however, been found empirically; these are discussed in the section dealing with design. free flight of heavier-ihan-air craft—the motions of aero- planes under the influence of air reactions and gravity have been extensively studied. an aeroplane is a body upon which the air can be made to exert a large force nearly perpendicular to the direction of motion. the magnitude of this force (r) can be represented by the equa- tion rk =kgpv°*s where s is the wing area. &, is here a function of the shape and attitude of the aeroplane®; it has a maximum value occurring at incidences in the neighbourhood of 18°..° when flying at a given speed in air of given density, the pilot, who has control over incidence, can thus call into play reactions up to a definite limit- ing valuc, but no greater; if he increases his incidence beyond that which gives the maximum reaction his aeroplane is said to be stalled. the principal use of the reaction r is to balance the weight of the aeroplane in steady straight flight, but it is also used to generate the large accelerations which are required in a rapid turn at high speeds. 6 ratio of span to chord. 7 rtl is here defined as ae where s is the wing area. pvr 8 a@ is the incilence of a wing of aspect ratio a required to give the same lift coefficient (ki) as a two-dimensional wing at incidence o al, alo is the drag coefficient of the finite wing and no is the ed coefhcient of the two-dimensional wing, when the lift coefficient is kk. ®°variations of kr with change of reynolds number are here neglected. | it is at this incidence that ‘‘ dead air’ forms behind the wing, as described in the previous paragraph. aerodynamics {f the speed falls so low that the maximum value of r is less than the weight of the aeroplane, steady flight becomes impossible, so that there is a minimum speed of steady flight for a given aeroplane in air of given density; this is called the stalling speed and is equal to “kr(max.) where w is the loading of the wings per unit area. in the absence of any propulsive force, such as that provided by an engine and airscrew, the reaction (r) is always inclined back- wards, through some angle y, from a plane perpendicular to the direction of motion, hence steady horizontal flight is impossible without the expenditure of power, and in the absence of such power the acroplane must descend along a path inclined below the horizon- tal through an angle equal to y. when the aeroplane flies in a curved path, the reaction (r) is re- quired both to provide the necessary accelerations and to balance the attraction of gravity. the magnitude and direction of the re- action may now be expressed in the form mg’, where #7 is the mass of the aeroplane and g’ is the vectorial sum of the gravitational field and the inverse of the accelerations of the centre of gravity.! g’ is described as the “ apparent gravitational field,” because the sensa- tions of the pilot are those which he would experience in a ficld of this magnitude and direction. it is usual to adjust the aeroplane so that its symmetrical plane is parallel to g’, when it is said to be cor- rectly ‘‘ banked ”’; the stresses in its main structure are then approxi- mately g’/g times those in straight flight. the magnitude of g’/zg has been measured in many manoeuvres, a value 3 is common in violent manoeuvres, values greater than 7? have been recorded for very short periods, but values in excess of 4-5, continued for more than a few seconds, have been found by independent investigators to cause temporary loss of sight and other faculties. when g’ is greater than g the acroplane will stall at a higher speed than in straight flight, though at the same angle of incidence. if the aeroplane is to be in equilibrium, the reaction (r} must pass through its centre of gravity. if it docs not do so, angular accel- crations will occur. ‘the moments of inertia of the aeroplane about axes through the centre of gravity are in general so small that these angular accelerations may be very large, so that rapid rotations would be quickly generated were it not that so soon as appreciable angular velocities occur, additional moments are generated which tend to stop them. the limiting rate at which an aeroplane can be rotated is thus determined by equating the moments which can be brought to bear about the centre of gravity, to the opposing mo- ments generated by the rates of rotation. the three control sys- tems, by which moments about three axes through the centre of gravity can be generated, are described in the article aeroplane, when an acroplane is disturbed from an attitude of balance, moments will be generated even when the “ controls ”’ are not moved; it is important to determine whether or not these will be of a kind to restore the equilibrium attitude. if the aeroplane be imagined suspended in an air current, so that it can rotate about a fixed axis through its centre of gravity, then it is said to be “statically stable”? about that axis if any small rotation results in a moment which opposes the rotation. to achieve “ static’ stability in conventional acroplanes, it is usual to fasten certain fixed areas, called the tail plane and the fin,’ to the end of the body, at some distance behind the centre of gravity; these act in the same manner as the feathers of an arrow. “ stai- ic ” stability can be increased cither by enlarging these areas, by increasing their distance behind the centre of gravity, or by moving the centre of gravity forward, in relation to the wings. stability.—in free flight an acroplanc is said to be completely stable when, however it may be disturbed, it will eventually sctile down to steady straight flight on an even keel. it is pos- sible to make an aeroplane which will do this, but complete stability does not necessarily follow from the possession of “static” stability. a statically stable aeroplane, if disturbed from steady straight flight on an even kecl by a motion in the vertical plane containing the direction of flight, will gencrally execute an oscillation of long period—z0 sec. or more— about the equilibrium path, and this oscillation may increase or decrease in magnitude in accordance with factors which may be independent of the static stability. in the majority of aero- 1 this procedure may be expressed by the statement that the re- sultant force upon the aeroplane is obtained by combining the * 7! * « weight mg with the centrifugal force os where 7 is the velocity and r the radius of curvature of the path of the acroplane. with this notation 9’ =eg?+ (=) : 2n.a.c.a. 203. 3 see aeroplane. 27 planes such oscillations decrease, but occasionally, particularly in models, an increasing oscillation is observed. the principal factor which causes an increasing oscillation is a large longitudi- nal moment of inertia. again a statically stable acroplane, dis- turbed laterally from steady straight flight, may execute a slowly increasing banked turn, requiring however many seconds for the increase to be serious. this form of instability is intensified by high directional static stability, but can be eliminated by a sufficiently large dihedral angle4 the study of the complete stability of an aeroplane for small disturbances from steady flight has been taken up by bryan, bairstow and others, who have reduced the mathematics to a tractable form. it is not essential for successful flight that an acroplane should be completely stable, nor even statically stable; unstable acro- planes have been very successful, particularly as fighters. much misconception has existed in the past as to the relation between stability and safety. the danger of flying, apart from breakage or fire in the air, is almost entirely associated with the manocu- vres of getting off and landing, when the desirable quality is con- trollability rather than stability. provided that static instabili- ties are not too great, the disturbances to which they give rise in normal flight are slow enough for the pilot to deal with them instinctively and without serious strain, and the “question has been much debated amongst experienced pilots, whether or not an unstable aeroplane is easicr and safer to bring to earth than a stable one. an advantage of complete stability is that the pilot, in a long straight flight, can remove his attention from control, and either rest himself or attend temporarily to other duties, but this ad- vantage has been largely offset in the past by the fact that it is hard to balance an aeroplane so accurately that it will not then fly on a slightly curved path. this difficulty can now be over- come by the use of a gyroscopic control on the rudder, which keeps the aeroplane on a given course; with the help of this device a stable acroplane can be made to fly for many miles on a straight course without any attention whatever from the pilot. even should the power to fly without human control be unim- portant, the designer should, however, know the degrees of static stability which his acroplane will have, for its characteristic behaviour in the hands of the pilot depends upon this factor. it is becoming the practice to design aeroplanes to have a small measure of static stability both longitudinally and directionally, but over-stability may be as bad a fault as instability. control problems.—the control of the acroplane by the pilot involves problems of great complication. apart from the needs of war, powerful control is required mainly for the purpose of manocuvring for a landing in a restricted area; these manoeuvres are generally carried out at the slowest practicable speed, when the acroplane is as near the stalling point as the pilot thinks safe, and since the pilot may accidentally stall his aeroplane in these circumstances it is of great importance to give him ade- quate control when flying at incidences both less and greater than the stalling incidence. unfortunately certain violent forms of instability may occur when flying above the stalling incidence, and considerable difficulty is experienced in providing means whereby these can be controlled. this whole matter has been studied extensively during the last few years and some progress has been made towards an understanding of the problems and the elimination of the dangers involved. bibliography.—in great britain, the united states and ger- many, the principal technical papers have been published in series of reports under the following titles: british, reports and memoranda of the aeronautical research commitiee (quoted belowas r. and ai); american, reports of the national advtsory committee for aeronau- tics (n.a.c.a.); german, technische berichte der flugzeugmeisteret. references.—theory of similarity: g. p. thompson, applied aerodynamics, ch. 11 (1919); l. bairstow, applied a erodynamics, ch. 8 (1920); lord rayleigh, r. and m., no. 15 and 39 (1910); see also r, and m., no. 900 (1924). theory of wings: f. w. i.anchester, aerial flight, aerodynamics (1907); l. prandtl, ‘‘ applications of see agi ee nd ey eee oe ee ee ne 4 see aeroplane. ’ these remarks do not apply to stalled flying in which violent unstable motions may occur, which are definitely dangerous. 28 iiydrodynamics to aeronautics ” with references to chief german authorities, n.a.c.a., no. 116 (1921); m. m. munk, n.a.c.a., no, 121, 142, 151, 191; j. s. ames, n.a.c.a., no. 213; glauert, r.and m., no. 723, 767, 824, 889, 910, 911, 946, 980; bryant and wil- liams, zrans. koy. soc. lend., scries a., vol. 225, pp. 199-245, and phage and simmonds, 7):d4., pp. 303-330. equilibrium, stability, and control: f. w. lanchester, aerial flight, aerodonetics (1908); g. hi, bryan, stedtlity in aviation (1911); g. p. thompson, applied aerodynamics (1919); l. bairstow, applied aerodynamics (1920). : (b. m. ? ‘ i. j.) aero engines.— all aero engines in use up to the present time work upon the four-stroke constant-volume-ignition cycle and the thermodynamic theory of them is similar in all essentials to that of any other petrol engine. (see internal combustion engine.) i. elements of the problem the great problem with which the designer of aero engines is faced is reduction of the ratio weight to horsepower, or as it is customarily expressed the ‘‘ pounds per horsepower,” to the smallest possible value consistent with reliability. but pounds per horsepower in the air cannot be taken simply as a figure obtained on the test-bed; an aeroplane must carry its own fuel, so that economy of fucl consumption is also of primary impor- tance. if we-take, in order to fix our ideas, the example of a 450 h.p. engine which weighs itself, say, goo lb. then the weight of fuel used per hour would be of the order of 225 lb., depending of course at what height the flight was made. if we consider, not the weight of the engine alone, which is 2 lb. per h.p., but the weight per ii.p. of the ‘* power unit (engine and fuel) for a four hours’ flight,” then the value of this when the aircraft leaves the ground will be not 2 but 4, and it becomes at once clear that the effective weight to power ratio of an engine in flight will be materially affected by its fuel economy. in 1915 there was no water-cooled engine which weighed, complete with water and radiator, less than about 4 lb. per h.p., and no air- cooled engine of less than 3 jb. per horsepower. in the decade 1915-25, these figures have been reduced to the neighbourhood of 2 and 1-6 respectively. the greater net weight of the water- cooled type is discounted, especially for long ilights, by better fuel economy. : tendency of destgn.—for an engine of given size (bore and stroke) both the power and economy are increased by an increase of the compression ratio. the tendency of design has thercfore been toward the use of higher and higher ratios of compression: from the figure 4-5 of the average motor-car the compressicn in aero engines has been pushed up to 6 to 1 and higher. ‘the limit to which it can be raised depends largely on the physico- chemical properties of the fuel available. with some fuels, such as benzol, an engine of 7 to 1 compression ratio will run quite smoothly. with others which differ only in chemical analysis and not at all in appearance, volatility or specific gravity, it would be impossible to run the engine at all. owing to the high degree of compression of the combustible gas, the ignition, when the spark is passed at the sparking plug, would take place with such violence as to cause ‘‘ detonation ” or “ knocking ” in the cylin- ders. in order that the compression ratio may be pressed to the highest possible point, therefore, the supply of suitable fuels for aero engines is a matter of great importance. the factor of head resistance-—the aero engine designer must also bear constantly in mind the factor of “‘ head resist- ance.” when an aeroplane is flying level, the whole available power of the engine is used up in overcoming the “‘ drag ” on body and wings as the machine tears forward through the air. in practically all machines nowadays the engines are at the front of the body or fixed between the wings, and if the drag is to be kept down, the area presented by the engine and acccs- sories, when viewed from the front, must be as small as pos- sible. a most important accessory, in this connection, is the radiator necessary with water-cooled engines. ii. evolution in design bearing in mind the four considerations of power, weight, fuel economy and head resistance, their eflect upon aero-engine aero engines design will now be traced. the designer must be a master of compromise. compactness, rigidity, revolutions, inertia forces, piston speed and area, bearing loads, valve area, effective cool- ing and the rest . . . each member of this complex company must be harmonised with all the others and at the same time consideration of each must be pressed to the farthest limit in the service of the two great aims of power and lightness. it is appropriate in the first place to point out the main sub-division into water-cooled anid air-cooled types. arrangement of cylinders-—-with our primary concern for weight to power ratio the advantages of the air-cooled type are not far to scek; there is the clear gain in weight of all the circu- lating water, besides the elimination of the weight and drag of the radiator. but the air-cooled engine designer cannot have it all his own way. for minimum head resistance of the engine itself a design of the “ cylinders-in-line ” type with the crank- shaft fore-and-aft will clearly have the advantage; but if for cooling we rely wholly upon a high velocity airstream it is imperative that no cylinder should be seriously more screened from this airstream than any other. ilence, although a few air-cooled engines of the cylinders-in- line type have been built, the two types which have been widely successful have been those in which the cylinders are arranged radially in one plane, or sometimes in two parallel planes, per- pendicular to the axis of the machine; so that all cylinders share equally in the cooling air. such engines, whether of the “ rotary ” or “ static radial ” types, present a large frontal area and offer very serious head resistance. water-cooled engines the power developed by an engine depends upon the mean piston speed, the total piston area and the mean-effective pres- sure in the cylinders. as to the first two the question at once arises: “‘ shall the requisite piston speed and area be achieved with a few large or many small cylinders, and what in each case are the limiting factors in regard to speed?” . cylinder design —the answer given by designers has been “many and small”; largely on account of the resulting com- pactness and of the rigidity obtainable with light construction. a long stroke means a heavy crankshaft and a cylinder reaching out a long way from the shaft centre-line and upsetting the slip stream from the propeller. for getting adequate piston speed, therefore, the tendency is toward a short stroke and high revo- lutions. large-bore cylinders in line, again, would lead to a long crankshait, involving both weight in itself and in the resulting crankcase; besides the danger of non-rigidity in a big engine of light crankcase construction. a further advantage of small cvlinders 1s the reduced danger of distortion due to uneven heat- ing while in service. with many small cylinders in line we get less piston area to the same length of crankshaft, but it is possible to make two or more cylinders operate on one crank and so the balance is re- dressed. considerations of this kind have led to the develop- ment of the “‘ vee ” and “‘ broad-arrow ” types of engine with two and three banks of cylinders. fig. 1 on plate shows one of the most successful modern water-cooled engines of the vee type, the curtiss d.12 which develops 400 b.h.p. at a total weight of 724 pounds. this engine has a crank throw of only 3 in. and the compactness of the resulting design is evident. a short stroke means high revolutions to get adequate piston speed, and this in turn involves big inertia forces (since they increase as the square of the revolutions) which become most scrious in their effect on the bearing loads at the crankpins. the limit to the closeness of spacing of the cylinder centre lines, with these high-speed engines, is largely a matter of the provision of adequate lengths for the main bearings and for those between connecting-rods and crankpins. valve problems.—the combination of high piston speed with good volumetric efficiency involves ample valve area and the freest possible form of induction system. this, combined with the need for a compact and symmetrical combustion chamber to diminish detonation, has resulted in the almost universal aero-engines and aerial navigation plate a a ae i. fic. 1, curtiss d-12 engine, of yoo b.h.p., weizhi 724 lb. fic, 2. jupiter air-cooled raha engine, 450 h. p. , weight 730 "9 (fig. 1. courtesy of ** aero digest”) ee ey 8. - 7 ’ lf e , => ¥ i'ic. 1. view of cockpit cof medern aeroplane, from the observer's seat. (roval air force photocraph. crown copvricht.) aero engines adoption of four valves in the cylinder head, two inlet and two exhaust, the inlet being rather larger than the exhaust. by this arrangement an effective inlet valve area is obtained equal to about 20% of the piston area. the maximum permissible gas velocity through the valve is about 170 ft. per sec. if good volumetric efficiency is to be maintained, and this therefore, with the valve area available, limits the mean piston speed to about 2,o00 ft. per minute. next to the big-end bearing loads, valve operation is most critically affected by high revolutions. to get sufficiently rapid opening and closing of a poppet valve for an engine speed of 2,500 r.p.m. the accelerations involve very heavy springs and terribly severe stresses in the cams and in the valve spindles. this difficulty of high-speed poppet valve operation makes the sleeve valve so attractive a possibility. fuel distribution—with a number of cylinders in line the equal distribution of the fuel between them is a difficulty. two or more carburetters must be used and the cylinders drawing from each carburetter must be arranged, so far as possible, to do so at equal time intervals. no one arrangement has been stand- ardised as the best: on most 12-cyl. vee engines one carburetter serves one whole bank of cylinders, but another arrangement divides the engine into two groups of the front six and rear six cylinders, and each group is served by one carburetter. in at least one design some weight is sacrificed for the sake of good distribution in the provision of a carburettcr to every two cylinders. air-cooled engines the two main classes of air-cooled engines are the rotary and the static radial types. in each the cylinders are arranged radially in one or two planes round the circumference of a circu- lar crankcase. thus all cylinders are satisfactorily exposed to the cooling airstream, but at the cost of presenting a large fron- tal area. the rotary engine, in which the cylinders and crank- case rotate with the airscrew, besides the fore and aft airstream, gets the extra cooling due to the rotational speed of the cylin- ders, but at the expense of considerable waste of power in air friction. | kinematically the two types are identical. in each there are two parallel axes: viz., that of the main shaft, round which the cylinders are symmetrically arranged, and that of the crankpin, to which the connecting rods are attached. in a rotary engine the crankpin axis is fixed and the cylinders rotate with the pro- peller, while in the static radial type the cylinders and crank- case are fixed and the crankpin axis rotates round the main shaft in the usual manner. the speed, and therefore the power, of the rotary engine is limited to about 1,200 r.p.m. by the enormous centrifugal forces on the rotating cylinders, and in consequence recent develop- ment of air-cooled engines has been wholly along the line of the static radial. one of the chief attractions of this type is the possibility of a very compact and rigid crankcase construction. this takes the form, very roughly, of a drum, with cylinders fixed radially round its circumference. only one crank is employed for all the cylinders which lie in one planc; one piston having a connecting rod with a “ big-end ” as ordinarily understood, while all the other pistons have ‘articulated ” connecting rods, which are attached by pin joints around the circumference of the big-end of the one master-rod. the loading on the one crankpin is very severe and special designs of big-end bearing are necessary to stand up toit. one of the most successful is the “ floating bush ” type in which a light bronze sleeve, free to rotate, is interposed between the crankpin and big-end bearing surfaces. a bearing of this kind, of course, is only possible with a crank of the “ buit- up ” type. valve design.—valve opening is contrived by push-rods and rocker-arms which are carried on the cylinder heads; the rods being operated fram a cam ring which is driven at the speed appropriate to the number of cylinders by an epicyclic gear on the main shaft. since the cylinders of a large, air-cooled, radial 29 engine may undergo an expansion while running of as much as 70/100oths of an inch, which is not shared by the push-rods, very careful design and proportioning of the rocker-arms are necessary if valve operation is not to be upset. fig. 2 on plate is an illustration of the jupiter engine, one of the most successful of the air-cooled radials. its nine cylinders, 5ix7 in. bore and stroke, are arranged in one plane and the engine develops 450 ii.p. at its normal speed of 1,700 r.p.m. and at a weight complete of 730 pounds. ill, materials with the cutting down of weight and the ever more severe conditions as regards temperature and compression ratio the problem of design is bound up with that of materials. aluminium alloys.—the introduction of the light alloys of aluminium was the first great advance in regard to materials. it is now possible to make extremely complex castings and drop forgings for a complete crankcase in this material, and, even before this had been successfully achieved, the introduction of the aluminium piston had largely reduced the mass of recipro- cating parts and so brought relief in crankpin bearing loads due to inertia. the high conductivity, too, of aluminium as com- pared with cast iron has made for cooler pistons and less tend- ency to carbonisation of the lubricating oil. the same property has led to extensive use of aluminium for making the cylinder heads of air-cooled engines, and, in combination with a steel liner, for complete cylinder construction of water-cooled types. it has even been tried to run a steel or iron piston direct in an aluminium cylinder. the two critical points in regard to high temperatures in a high-duty engine are the centre of the piston, and the exhaust valve, if this is of the poppet type. exhaust valve materials —exhaust valves are called upon to run continuously at a bright red heat and to resist scaling or burning under these severe conditions. the only steels which can successfully do this are certain of the nickel and chromium alloys and these in their turn introduce new problems: they are much more troublesome to machine than carbon steels and also they develop what is known as “ temper-brittleness,” a con- dition brought about by the high temperatures at which they are used, and their subsequent slow cooling after use. alloys of magnesium with copper and zinc are both lighter and stronger than the aluminium alloys. they cast well and machine easily and have been used to some extent for crank-cases with a saving of weight of about 35%. their disadvantages are a cost of about double that of aluminium and excessive liability to corrosion. iv. the special problem: altitude effect on engine and :airscrew—under comparable conditions the pressure developed in the cylinder of an internal combustion engine will be proportional to the weight, and therefore to the density, of the combustible mixture drawn in. at heights above ground level the atmospheric pressure falls off at a rate of roughly 1 in. of mercury per 1,000 feet. at the same time the average temperature of the air becomes lower at a rate and ina manner which varies from day to day. the density of the air will therefore fall off less rapidly with height than the pressure, and the relation between density and height may be very variable. it is now possible to measure the indicated power of an engine in flight, and experiments on these lines have shown that the indicated power is very nearly proportional to atmospheric density, provided a proper air-fuel ratio is maintained. this, however, is a matter of some difficulty. the “ depression ” produced in the choke tube of a carburetter and hence the quantity of fuel normally supplied through the jets will, for the same engine speed, be proportional to the square root of the air density. the weight of fuel supplied by a carburetter will therefore fall off less rapidly, as an aircraft rises, than the weight of air drawn in, and the mixture will become too rich. to com- pensate for this an “altitude control’ must be fitted which usually consists of a connection between the float chamber and 230 a point of low pressure in the choke tube. by the gradual open- ing of this passage the flow of petrol is reduced and the correct mixture maintained. engine and screw.—the useful performance of an aircraft engine cannot be considered apart from its airscrew. these two must be regarded together as the power unit, for the power and efficiency of an engine may be largely thrown away if combined with an inefficient airscrew. moreover, the efficiency of the airscrew is affected by the revolutions maintained by the engine. since both indicated engine power and airscrew resistance are proportional to air density, at constant speed, it follows that they both fall off at the same rate as the aircraft climbs. the b.h.p. of the engine, friction losses being approximately con- stant, will fall slightly more rapidly; but the difference is not large, except at great heights, and since airscrew resistance is also proportional to the square of the revolutions the difference between available engine torque and airscrew resistance will produce only a slight drop in the speed of rotation. a fixed bladed airscrew, therefore, combined with a normal engine will, as the machine climbs, continue to work at nearly constant revo- lutions and efficiency. variation of b.h.p. with altitude -—although the indicated power falls off, the mechanical losses in the engine as a whole will not vary much as a machine rises, so that b.h.p. may be expected to fall off more rapidly than indicated horsepower and air density. to determine by direct observation just how fast the b.h.p. falls off with increasing altitude in any given case would be a research of the highest interest. its accurate deter- mination would involve the two measurements of the speed, and the torque actually delivered to the airscrew. several at- tempts have been made to design a torque meter for use in the air, but the mechanical difficulties are so great that so far these attempts at direct measurement have been only par- tially successful. the alternative is to deduce how the b.h.p. varies from obser- vations of the performance of the aircraft. this again is a difli- cult calculation to make with certainty from figures for speed and rate of climb, since the necessary reduction from actual observations must rest upon an answer to the question we are asking: how the altitude has affected engine b.h.p. and airscrew eficiency. it is probably impossible, by tests in the air, to arrive at any satisfactory general law for diminution of engine b.h.p. with height, applicable to all engines, for the simple reason that the mechanical efficiency, which must always vary from engine to engine, and may even vary from day to day with the same engine, has so large an elfect on an aircraft’s performance as to produce very discordant results. take for example an engine giving 500 i.h.p. on the ground, and assume that the l.h.p. will fall off in proportion to air density and the friction losses remain nearly constant. if we calculate the b.h.p.at 20,000 ft. where the [.h.p. is halved, this will be 200 or 175, a difference of 123%, according to whether the mechanical efliciency is 90% or 85% on the ground. even 1°% variation of mechanical efliciency would have an appreciable effect on performance. the position may be summed up by saying that although such direct observations as have been made indicate that the b.il.p. falls off nearly in proportion to the reduced pressure, irrespec- tive of temperature variation, nevertheless the assumption made above is sufficiently close to justify its use in the prediction of aircraft performances considering the inevitable uncertainties of individual tests. this assumption makes the b.h.p. depend upon the density (and therefore temperature), diminishing in a linear manner with it, but more rapidly than the density itself. muinienance of ground power at alttiude-—since the power delivered by an engine will be roughly halved at a height of 20,000 ft., any plan for counteracting this effect of reduced air pressure holds out great opportunities for improved performance. but here the airscrew problem comes in. if the ground power of the engine should be delivered with the airscrew in low density air, then the engine will race away and burst, for unless the in- clination of the airscrew blades can be altered the resistance to rotation must fall off with the air density, the usefulness of aero engines maintaining ground power at altitude is therefore entirely dependent upon our being able so to vary the pitch of the air- screw blades as to prevent excessive engine speed and at the same time retain a good airscrew efficiency. this problem of providing a variable pitch airscrew is one which is at present occupying our designers and cannot be said yet to have been satisfactorily solved. supercharging for maintaining ground power—to maintain ground power, combustible mixture must be compressed so as to be delivered to the cylinders always at normal atmospheric density whatever the height. this would mean, at 25,000 ft., compressing in the ratio of 2-7 to r and then cooling through about 60°c. the necessary cooler involves additional head resistance and, if a piston compressor is used, the displacement will have to be about three times that of the main engine. all this means considerable additional weight, and every added pound is a dead loss at low altitudes and a serious handicap in getting off the ground. for this reason very high-speed rotary blowers offer the best chance of success. such a blower might either be gear-driven from the engine shaft, or some of the energy of the exhaust gases might be used through the medium of an exhaust-gas turbine. the difficulty attending a blower, if gear- driven from the main shaft, is that, with the exceedingly high gear ratio necessary, about ro to 1, the inertia of the blower rotor is enormous, and unless some slipping clutch device is provided the gears will be stripped on the very sudden starting up of the engine. the exhaust turbine drive is very attractive, for, whereas with a gear-driven blower the entire horsepower needed for com- pressing the air has to be supplied by the main engine, this type makes use of what would otherwise be largely wasted energy. the difficulties are chiefly those of providing turbine rotor blades which will stand the constant exposure to the hot exhaust gases and at these temperatures will retain a strength adequate for a speed of rotation of 2,500-3,000 revolutions per min- ute. scarcely jess serious is the problem of a shaft-bearing at these speeds when the high temperature makes oil lubrication impossible. v. minor problems wireless interference-—while the engine designer has been improving the power and efiiciency of his magneto and ignition system to suit high-compression engines, great strides have been made by the wireless branch in short wave-length instruments which can do without the trailing aerial. —the magneto and high- tension leads produce waves which upset the sensitive wireless sets, and methods must be devised of so screening off the effects of the ignition system as to allow advantage to be taken of the improvements in signalling apparatus. the high-tension leads are therefore enclosed throughout their whole length in a woven metal sheath which is “ earthed ”’ to the engine body and at the sume time the plug-heads and contact-breaker are enclosed in small metal cowls which are likewise earthed. but the hard-won improvement in ignition efliciency is cut down by the added electrical capacity introduced by the new screening arrange- ments, the effect of which, and the additional weight involved, are a dead loss to the engine designer. oil temperature control-—the functioning of a system of forced lubrication, in which oil is delivered by a pump to all parts of the engine, is very greatly affected by the viscosity of the lubricant. since the viscosity of the lubricating oil will fall to about 1/roth of its value for a rise in temperature from 40° c. {to 100° c., some control of oil temperature is necessary. during the cold period at starting, excessive pressures will be developed by the pump; and although these can be dealt with by relief valves, the flow to the bearings will be very small, and a long period of warming up is necessary until the flow is sufficient to allow full loading to be imposed on the bearings. when flying at low altitudes and in warm climates, and more especially with airship engines which are usually not open to any cooling air- stream, it is necessary to provide for cooling of the oil aiter it has left the engine and before it is redelivered by the pump. aero engines the problem is a peculiar one, for although the amount of heat to be abstracted is comparatively small, the rapid rise of viscosity as the oil is cooled has the effect of producing a thick coating of viscous oil on any surface where cooling is effective. the conductivity of this viscous layer being very bad, the re- mainder of the hot oil passes on out of the cooler with a very inadequate reduction of its temperature. the problem is to prevent the oil, as soon as it is cooled, from remaining in contact with the cooling surfaces and so preventing the cooling of that which comes after. the problem will not be solved until a cooler has been designed which is light, which cools effectively and can deal with the high pressures developed while the oil is cold. water recovery.—a problem peculiar to airships is that of compensating for the weight of fuel used during a long flight. in flying to egypt a large ship would burn some 25 tons of fuel oil, and unless this weight can be made up in some way an equi- valent quantity of hydrogen will have to be allowed to escape. an alternative is, while drawing liquid fuel from the tanks, to use less of this and to make up the balance of fuel required by drawing from the gas bags the corresponding weight of hydrogen necessary to maintain balance in regard to buoyancy. research upon the use as fuel in an engine of a mixture of oil and hydrogen has yielded most promising results and this may prove the most workable solution of the problem. the alternative of maintaining the dead load on the airship by condensing and storing the water in the exhaust gas from the engines is a matter of considerable difficulty, but it should not prove insuperable. in the combustion of a hydrocarbon fuel the water formed is 9 times the weight of the contained hydro- gen. now an oil fuel contains about 14% by weight of hydrogen, so that in addition to any moisture carried into the engines with the air there will be 9x-14(=1-26) lb. of water for every pound of fuel burnt. this is by no means all recoverable, but with a water recovery apparatus capable of cooling the exhaust gas to within 10° c. of whatever may be the surrounding atmospheric temperature and of collecting all the water formed, the possible recovery should be from o-9 to 1-o times the weight of fuel burnt. vi. future developments increase of size v. increase of gas pressures.—apart from piston area and piston speed the only way of increasing the power of an engine is to increase the mean effective pressure. this can be done either by raising the pressure or lowering the tempera- ture of the combustible mixture drawn in. the latter alternative has obviously only limited possibilities, but within these limits the use of fuels with high latent heat of evaporation, which therefore give more cooling of the charge, has been successfully applied in racing motor-car practice and may be also in aircraft engines. to raise the pressure of the mixture drawn in means the provision of a blower, not only to maintain ground pressure at altitude but to raise this pressure throughout above atmos- pheric. under such conditions an engine of the same bore and stroke would have to have a heavier crankshaft, connecting rods, pistons, etc., to stand up to the larger gas forces developed, the necessary blower and driving gear will be larger and heavier, the air coolers will be heavier and give more drag, so that it might be suggested that the extra power could be got at once by means of a bigger engine (7.e., by more piston area), while leav- ing the gas pressures the same. only the detailed calculation of weights and estimates of drag in the two cases can settle the question of which plan will produce the most effective engine, but the small engine with high gas pressure starts with an initial advantage so far as its own head resistance is concerned. the two-stroke cycle-—to design for a working stroke every revolution is not the easy way toward reducing weight per h.p. which at first sight it appears. the getting of the charge into the cylinder with only the limited area available for inlet valves is, at high speeds, one of the limiting conditions even when the time of half a revolution is available. if it has to be accomplished in one-third of the time, or less, the problem is proportionately greater. any lowering of engine speed, if combined with all the 3i extra weight of the necessary compressors for scavenging and delivering the fresh charge, may very soon bring down the extra horsepower per pound from a two-stroke cycle to a very narrow margin. it is unlikely, moreover, that the same thermal efh- ciencies will be achieved as with the four-stroke cycle. never- theless the two-stroke cycle is a possible line of advance, more especially with the greater perfection of compressors for super- charging and in combination with the diesel type of fuel injection. sleeve valves.—one of the most difficult aero engine problems is the satisfactory operation of poppet valves at high speeds and high temperatures. the drawback to the use of sleeve valves has been in the past a tendency toward gas leakage, excessive oil consumption and large mechanical loss due to friction; but with the single-sleeve valve it is possible, by skilled design to minimise these troubles, and this type of valve holds out ex- tremely attractive possibilities. the absence of all springs in the first place and of the excessive accelerations which are so detrimental to cams, would at once eliminate a constant source of trouble. moreover the sleeve would necessarily be cool, so that tendency to detonation would be reduced; further, it would be possible to provide ample and unobstructed port area so that volumetric efficiency could be maintained at high speed, for apart from the absence of obstruction due to poppet-valve stems in the port, with sleeve valves we are no longer limited to the cylinder end in providing valve port area. the compound engine forty per cent or more of the heat energy of the fuel is at present thrown away in the exhaust of a high-duty engine; where weight is no consideration some of this can be used for raising steam, as in the still engine designs, but such methods of improving efficiency are probably out of the question for aircraft. the question remains, ‘“‘can some of this waste heat be saved by expanding the hot gases much further in a low-pressure cylinder?” the difficulty lies in the transfer of the hot gases at 1,000° c. or more from the high- to the low- pressure cylinder. so long as this has to be controlled by poppet valves the difficulties are probably insuperable; no valves could be expected to stand up to such treatment, when it is remem- bered. that they would need to be operated from outside, and that therefore there would be continuous leakage of hot gas down the valve stems where they slide in the guides. the only hope for the compound engine seems to be through the develop- ment of the sleeve valve. it is a possibility, but much research will be required before the difficulties inherent in such a scheme can be met or their magnitude even satisfactorily settled. the diesel engine.—the use of this engine is hardly more than a pious aspiration so far as aircraft are concerned, in spite of its manifest advantages. not only would it give a very high fuel economy on account of its high compression ratio of 12 or 14 to r instead of 5 or 6; but it would burn a fuel of one-third of the price of petrol to-day, and all the weight and complication of magnetos and sparking-plugs would be swept away. there is no fundamental reason why the advantages of compression igni- tion should not ultimately be achieved without excessive weight; but since present-day aero engines can be built at a weight of less than 2 1b. per h.p. and the corresponding figure fora marine diesel engine is of the order of 30 1b. per h.p., it is clear that a jong road of patient research has to be traversed before a diesel type engine can be seriously considered for heavier-than-air machines. jet propulsion—more than one proposal has been made that the force for propelling an aircraft forward should be obtained by shooting out a jet of air behind on the principle of the rocket. to obtain efficiency, however, when using this principle, it would be necessary that the forward speed of the aircraft should be of the same order as that of the jet backwards, while at the same time the drag due to air resistance has to be overcome by the reaction between the aircraft and the jet. the two conditions are incompatible. but even under the best conditions an air- screw which is rotated by jets at the blade tips and eliminates other driving mechanism could never hope to achieve more than a very poor fuel economy when compared with a present-day heat engine. (reds) 32