{
    "system": "GoGuides Verified Text",
    "api_version": "verified-text-v1",
    "status": "ok",
    "response_type": "verified_text_record",
    "source_key": "britannica_1926",
    "source_title": "Encyclopaedia Britannica (1926)",
    "license_code": "public_domain",
    "attribution": null,
    "license_url": null,
    "chunk_id": "1926:asbestos:d0aef69ec4da",
    "title": "ASBESTOS",
    "section": null,
    "hash_alg": "sha256",
    "hash_sha256": "39b234dad9b455cce85643cdcf406bf55f59d363770c54fc25ad196019e894a0",
    "normalizer": {
        "name": "ggnorm",
        "version": "1.0"
    },
    "verified_text": "an incombustible, fibrous material usually found in serpentine formation in many parts of the world, with numerous varieties which are known or described as mountain flax, leather or cork, amianthus, tremolite, crocid- olite, and by other names. deposits in italy.—the asbestos from the val malenco or val tellina in northern italy, occasionally found in the form of thread from 3 to 20 ft. in length, is strong, tough, saponaceous, incombustible and free from objectionable impurities, and was first used commercially in 1871, when an asbestos factory was established in glasgow. a few years later another factory was established in turin. canada.—meanwhile other deposits of asbestos were dis- covered in the province of quebec. while the chemical proper- ties of this canadian, or “ chrysotile,’ variety are similar to those of the italian fibre, there is a marked difference in their physical characteristics; the canadian material, having a finer texture and being easier to obtain and to manipulate by machin- ery, came rapidly into extensive demand, although for some purposes italian asbestos was, and is still, considered more suit- able; on analysis it is shown to contain:— per cent lime and magnesia 37-84 silica . : 41-69 oxide of [ron 3-01 potash . 0-85 soda 1-4] alumina a ae s 2 2°37 moisture evaporated at 100°c. 3-04 loss on heating to white heat, water of hydration and organic matter . ; : ‘ . ' asbestos—-ashanti other sources —three other principal sources of supply were afterwards discovered and developed (a) in south africa, blue in colour and containing, when compared with italian and canadian asbestos, a much larger percentage (from 30 to 40% of oxide of iron and a correspondingly smaller percentage of magnesia; (b) in the ural mountains, similar to the canadian but a little darker in colour and slightly less pliable; and (c) in rhodesia. as this latter variety satisfactorily meets the re- quirements of manufacturers, its use is fast increasing. asbestos is now manufactured in england, in the united states, france, germany, austria, russia, italy, spain, bel- gium, canada and australia. the production of the raw ma- terial has sprung from about 200 tons (italian) about 1868 to a total from all sources of over 250,000 tons in 1924. the united states are the largest consumers, and it is believed that the asbestos factories there absorb over 70% of the output of the raw material from the canadian mines. lu ses.—asbestos is now made into a great variety of articles, used in practically every branch of trade, and in laboratories, surgical, medical and hospital equipment, no department of industry or scientific research to-day being complete without asbestos in one form or another. probably the most important development in the use of asbestos in recent years has been in connection with building operations. tiles for roofs, flat sheets for partitions, ceilings, etc., corrugated sheets for factory and other buildings, in which asbestos in combination with portland cement, and marketed in great britain and abroad under various trade names such as “ poilite,” “‘ everite,” “ t.t.,” “ eternit,”’ etc., has come into large and rapidly increasing demand in nearly all parts of the world. as the economic value of asbestos depends, not only on its power to withstand high temperatures, but also on its low thermal conductivity, and its partial resistance to the attacks of acids, its field of usefulness is practically unlimited. bintrocrarny.—a. l. summers, asbestos (1919); u.s. asbestos and mineral corporation, asbestos from mine to finished product (1919); m. a. allen and g. m. butler, asbestos (tucson, 1921); a. w. m. boram, asbestos and kindred minerals (1921); j. s. dillon, asbestos in 1919, u.s. geological survey (1921); u.s. bureau of foreign and domestic commerce, asbestos, world pro- duction and trade (1922); also ‘‘ the development of the asbestos industry 1871-1924\" in india rubber journal (london, aug. 2 1924) and asbestos (philadelphia, monthly). cleve fe) ashanti (see 2.724), a dependency of the gold coast colony, is divided into two provinces, the eastern with headquarters at kumasi, the capital, and the western with headquarters at sunyani. its area is 24,560 sq. m. and its population 406,594. king prempeh, who in 1806 resisted the establishment of a british resident at kumasi and was sent as a political prisoner to the seychelles, was allowed to return to ashanti at the end of 1924, although without special powers. the golden stool on which he was crowned was hidden in ashanti, and was only re- discovered in 1921. ashanti is administered by a chicf commissioner, assisted by provincial and district commissioners. although it is con- stitulionally separate and distinct from the gold coast colony and the northern territorics, all three are treated, so far as finance is concerned, as a single unit. no satisfactory estimate can be made, therefore, as to the revenue and expenditure, except those for purely local purposes. similarly it is impossible to state the actual trade statistics and the only satisfactory test is supplied by the tonnage figures of the railway, which show that 61,176 tons of cocoa, 2,983 tons of kola and 4,085 tons of other produce were railed to the ports in the year 1923-4. as in the gold coast colony proper, cocoa forms the prin- cipal item of commerce, and is the foundation of the peace and prosperity of the country; practically all the plantations are owned and managed by natives. agricultural stations have been established at kumasi, juaso and ejura, and special efforts were made during 1923~5 to encourage the planting of oil palms and the cultivation of coffee. gold-mining is of importance, although only two mines were producing in 1924. excellent mo- tor roads, 7214 m. in length, have been constructed, including ashfield—asiago, battle of one between sunyani and kumasi; the great north road from kumasi to ejura is the finest and carries a heavy trafhc to and from the north. a second railway from the gold coast to ashanti, linking up accra and kumasi, was completed in aug. 1923. the govt. maintains a fleet of motor vehicles at kumasi. brsitrography.—d. kemp, nine years on the gold coast (1898); c. hayford, gold coast native institutions (1903); frederic h. gough, 7 he ordinances of ashanti, etc., revised edition prepared under the authority of ‘‘ the reprint of statutes ordinance, 1909 (roto); s. r. b. a. ahuma, the gold coast nation and national consciousness (1911); l. p. bowler, gold coast palaver and life on the gold coast (1911); c. hayford, gold coast land tenure and the forest bill (1912); h. waetjen, zur geschichte des tauschhandels an der goldktiste um die mitte des r7ten jahrhunderts (1915); reports, notes of cases and proceedings and judgments in appcals, etc., and references under rules, orders and ordinances relating to the gold coast colony (1915); cc martin, les possessions britanniques en afrique occidentale. cete de or (renseignements coloniales, etc., 1917); t. w. h. migeod, “ tribal mixture on the gold coast, \" jour. african soc., vol. 19, pp. 109-125 (1920); sir f. c. fuller, a vanished dynasty: ashanti (1g21); r.s s. rattray, saa ie (1923). ty - ashfield, albert henry stanley, 1st baron (1874- ), british business man, was born at derby nov. 8 1874. he spent his early years in the united states, and was educated at american technical schools and colleges. he entered the service of the detroit city street railways and had a successful business career, becoming general manager of the company and subsequently of the public service corporation of new jersey. in 1907 he returned to england, and took up the position of general manager of the metropolitan district railway and soon after became managing director of the traffic combine which in- cluded the london underground electric railway companies and the london general omnibus company. in tror4 he was knighted. on the formation of mr. lloyd george’s govt. in evi ne ill, lly, wh ee ial! 239 1916 sir albert stanley was elected to parliament as a coali- tion unionist for ashton-under-lyne, being included in the cabinet as president of the board of trade. he was a notable instance of a minister selected as a “ business man” and not for any of the usual political considerations. he resigned from his office in may rorg, and in jan. 1920 was raised to the peerage. asiago, battle of, 1916.—the asiago plateau was the scene of various battles on the italian front during the world war (see italian campaigns); but the name of the battle of asiago was given to the fighting which took place on the tren- tino front during the austrian offensive of 1916. an attack from the trentino with the object of cutting the italian communications with the julian front, and so bottling cadorna’s main force in what krauss (in his book die ursachen unserer niederlage) calls “ the venctian sack,” was an operation which could not but commend itself to the austrian general staff. even falkenhayn, who refused his co-operation to the pro- posal made by conrad von hetzendorff in dec. 1915, admits that “it was very inviting.’ he did not, however, agree with conrad and with krauss, then chief of the staff to the archduke eugene, that a completely successful attack would have a deci- sive effect. he doubted the possibility of collecting the force he considered necessary for the enterprise (25 divisions), and did not think that the railway communications were adequate to supply such a force. he was, morcover, anxious about the rus- sian front. conrad’s plan was to attack through the asiago and arsiero uplands, in the direction of vicenza and bassano, and when he failed to convince falkenhayn that the effort should be a joint one, he determined to attack independently. cadorna’s plans.—cadorna’s line of argument, when rumours of attack began to arrive, resembled that of falkenhayn. he seat ma os a aa i ent 1 (le vit i dodi cbsa a | gm, sins ing gat t li kf f mn “ay cn bit y 4 0 ie ws “ y ah ty z yo © spitz. se cam fom phos lee] te ‘fe hide skate ssstnn fel na fee ay any ee wy aig. ~~ one \\ se = lene... 1 4s if . lec 7 a i ia ! — i ae s nae pan w a r h ii roe ar} xe 7 23 elt ti ne i ee oa nera et ~ t as j ep ; net a | uf ine o4e ng | es : lo ty, waar, ae he my rhea is xt ses ( ? an ae hens 1 + f orr ali ye [b/g ocis wn noe ore : cd gee cy ot oro ta rent a a hay lly viging ew schio pee p : ae > mig ‘ &: ak a as 7 ma’ ad om aa ne a oh cote portule yyy “ s ana y wal saa ya sth iff ‘hy ny al san ni, ze anes , ¢ a4 von a ‘ 4 oo e- : i ni a a a dn a em “2 win ly } ys 2 wal dinas ~ . = yu ayes? se ws an fe wh + 3m magnaboschi ” the 2 vif wty \\ se \\!' “, yy 4 2-9 : 4 il lid), - ia e rm. pah, wang al 9 , ; we pallet eae a halt. mig la ws ene tr sow ty / rgn & acs, soy r ° y fy i, aa, 1. tss : wy ok “a an wy ut 0 ra ar ame vs : fra a fi esltlly is swe . stas aa marostica | battle of asiago english miles 2 4 6 kilometres 240 did not think that conrad could spare troops for an offensive on the grand scale, and he was of opinion that the railway com- munications in the trentino were insufficient for such an offen- sive. he concluded that the reported movements in the tren- tino signified a limited attack, to be undertaken with the object of hampering his own offensive action towards the east. he had continually urged upon brusati, who commanded the i. army, that his rele was strictly defensive, now that his first duty, that of reducing the length of the trentino front and occupying strong defensive positions previously selected, had been successfully performed. on various occasions cadorna had emphasised the necessity of strengthening the positions chosen by him for defence, but his instructions had been insuffi- ciently regarded. when cadorna went to visit the lines in person, at the end of april, he found that while the front lines, in many cases unsuit- able for prolonged resistance, had been elaborately fortified, in various sectors the reserve lines which he had indicated as the “battle positions ”’ were almost untouched. cadorna ordered the positions to be modified, and the work of preparation was hastened on; but the enemy attack seemed imminent, and it was impossible to set about a complete re-organisation under the immediate threat. on may 8 brusati was replaced by gen. pecori-giraldi, the commander of the vii. corps (iii. army), and within a week the austrian offensive was launched. disposition of the forces—the austrian attacking force was arranged in two armies, one behind the other, dankl’s xi. army in front with 9 divisions, koevess’s iii. army in support, with 5 divisions. the troops in the val lagarina and val sugana were not included in this force, which was to make its offensive between the two valleys, where only supporting attacks were to be carried out. krauss, as chief of the staff to the archduke eugene, was opposed to the disposition of the two armies and to the limitation of the attack to the hill country. he urged that the front of attack should from the outset be divided be- tween dank! and koevess, and pressed for the concentration of attacking masses in the valleys, especially in the val sugana. but the original plan, prepared in all its details by conrad anc his staff, was not modified; it would seem that the archduke and his chief of the staff had little freedom of action. the tacti- cal direction of the attack was entrusted to dankl, who had at his disposal some 180 battalions. to meet the attack, pecori-giraldi had in line and immediate reserve, between lake garda and the val cismon (north of the val sugana), 130 regular battalions, 7 battalions of customs guards and 45 battalions of territorial militia, the latter at very low strength and of small fighting value. but his centre was weak, for 28 battalions of regular troops were in the val sugana sector, and on the actual front of attack he had only go regular battalions. another division was concentrating at desenzano, and 5 more were on the tagliamento ready to be sent in support in case of need. the artillery strength consisted of 851 guns, of which 348 were of heavy or medium calibre and 259 were light guns of position. dankl had, initially, a big superiority in infan- try, but his great advantage lay in his preponderance of artillery strength. between the val lagarina and the val sugana were concentrated some 2,coo guns, of which nearly half were of heavy or medium calibre, including 40 305-mm. howitzers, four 380’s and two or three german 420’s. the austrian altack.—the offensive opened on may 14 with a very heavy bombardment along the whole line from the val lagarina to the val sugana, but the concentration of fire was most intense between the vallarsa and the upper astico, and against this sector, the following day, the main infantry attack was launched. the plan was to attack first with the right wing of the xi. army, commanded by the archduke charles, sup- ported not only by its own artillery, but by flanking fire from the massed guns on the lavarone plateau. when the right wing had made sufficient ground the left wing was to come into action against the italian line north of the upper astico. for a time everything went well with the attack. the italians were driven back from their ill-chosen front lines, losing many asiago, battle of prisoners and guns, and by may 1o their position was very grave all along the line from the vallarsa to the astico. the retiring troops had failed to make a prolonged stand on the insufficiently prepared battle positions. on the left monte pasubio, the key position, was only lightly held by reserves, which had been hurried up in the nick of time, and in the centre the austrians had driven the defenders off the main line of defence, which ran from monte maggio by campomolon to spitz tonezza. the 37th div., which had held this line, had been forced back beyond the posina and the astico, and there were gaps both to right and leit of it. the austrian right was pressing hard on the italian main positions west of the vallarsa (coni zugna and passo buole), and was collecting forces to attack pasubio. there was breathing space for a moment in the centre, but the austrian left now came into action, kraut- wald von annan’s iii. (graz) corps being launched against the italian 34th division. ample italian reserves were now on the move, but it was a race. krauss blames the archduke charles for waiting with his xx. corps until the guns could be brought up to support a new attack, instead of driving through at once to arsiero with all available troops. the risk was not taken, and the short respite gave time to close the door in the face of the invader. the course of the battle, with the necessity of bringing up reserve divisions, led to a reorganisation of the attacking forces, koevess taking command of the left wing and dankl of the right. in the vallarsa and pasubio sector the attack developed strongly, but without success. farther north the archduke charles was wailing for his guns and reserves, and between his left and the iii. corps, kirchbach’s i. corps was coming into action. the iii. corps was hammering against the italian 34th div., which was not to resist for long. the situation in the centre was critical, and cadorna consid- ered that if the austrians were able to concentrate on the weak spot and keep up the impetus of their attack they might succeed in breaking through to the plain. on may 20 he went to udine, and after consultation with the duke of aosta and frugoni gave orders for the concentration of a reserve army in the venetian plain. the first four corps of this reserve army (the v.), which were made up of units drawn from the ii. and iti. armies, were ready on june 2. meanwhile the austrians were continuing their advance in the centre, but they could gain no ground against the italian leit. by may 22 bertotti’s 44th div., sent up from desen- zano, was in solid possession of both sides of the vallarsa road and of pasubio, and in touch with ricci-armani’s 37th div. on his left. it was in this sector that the austrian offensive met its fate. owing to the steadfast resistance of the troops under ricci-armani and bertotti, dankl could never secure a sufficient width of front for his advance. if the zugna ridge had fallen, the effect upon the pasubio position, already a salient, would have been more than serious, and upon the holding of the pasu- bio lines depended the maintenance of the positions to the east- ward. if pasubio went, the line south of the posina was turned, and the austrians had a new route to the plain by the valli dei signori, as well as the opening they were now making for, by the lower astico. till may 30 the attacks on the zugna ridge were continuous, but no progress was made. the attempts upon pasu- bio were as incessant, and lasted longer. the austrian infantry advanced along the great ridge from col santo; they came up from anghebeni and chiesa in the vallarsa and from the val terragnolo by the borcola pass. all efforts were in vain. northeast of pasubio, along all the rest of the mountain front to the rim of the val sugana, the austrians gained notable suc- cesses. koevess drove back the italians across the val d’assa, and thence still farther east, across the parallel valleys of nos and campomolon. to the southwest dankl’s left crossed the astico, and after heavy fighting pushed the italians back across the val canaglia, while his centre gained ground across the posina, south of arsiero. at the end of may the italian position still seemed critical, and cadorna gave orders for the with- drawal of stores and heavy guns from the isonzo front to beyond asia minor-—-astor, w. a. the sile, south of treviso. he believed he had the measure of the austrians, but he omitted no precautions. his confidence, in fact, was justified. the impetus of the austrian attack was dwindling. losses had been very heavy; the attacking divisions were beginning to lose their offensive value, and the reserves were insufficient. already on may 27 conrad had been compelled to ask falkenhayn to send a division of the austrian xii. corps, which belonged to prince leopold’s army group. and cadorna’s v. army was practically ready in the plain. on june 4 brusilov broke through at leeck. the first news of the russian attack did not perturb austrian headquarters, for conrad thought that his line in the east was firmly held. in a few days the situation was changed altogether. but even before the news of the disaster had reached bozen it was clear that the offensive against italy had failed. koevess was to gain a little more ground. by june 8 the austrians were only three miles from valstagna, low down in the brenta valley, but they had shot their bolt. south of asiago and south of the posina the attack was continued for 10 more days. here were the short- est routes to the plain and here the austrians had been able to bring up their guns in sufficient numbers. kirchbach made a great effort against the monte lemerle-monte magnaboschi line, while the archduke charles strove hard to win room south of arsiero by incessant attacks in the novegno sector. no further ground was gained. retreat of the austrians —by the middle of the month ca- dorna had begun the first move of a counter-attack, but the aus- trians were now getting ready to go out of the salient and back to a strong line which they had already selected. attacking on may 25, all along the line, the italians found the invaders in retreat. it was too late to develop the counter-offensive which was to have been directed against the two sides of the austrian salient, and cadorna relinquished the idea of a big attack as soon as he found a resistance which could only be overcome by long preparation and the use of artillery in mass. casualties on both sides were very heavy, and indicate the severity of the fighting. the austrian losses were estimated at over 100,000; the italian figures, up to the end of the counter- movement, were 35,000 killed and 75,000 wounded, with 45,000 prisoners, many of whom should be reckoned among the wounded. the austrian attempt to break through ended in definite failure but the attack was well planned and conducted with skill and determination. failure was due to the fact that conrad met with a resistance which went beyond his calculation. falkenhayn and cadorna had summed up the situation rightly. bibliography.—la guerra italiana nel 1916 (1917), from official documents; t. n. page, [taly and the world war (1921); l. cadorna, la guerra alla fronte italiana (1921). cw. mcc asia minor (see 2.757), a geographical term used both in a wider and a narrower sense. in the wider usage it covers asiatic turkey within the pre-war frontiers, excluding the arabian peninsula. in its narrower sense it means the anatolian penin- sula, together with a vague zone extending eastwards along the roots of the peninsula into armenia and kurdistan. as a result of the world war, asia minor in the wider sense has been broken up politically into the turkish republic and the three territories of syria, palestine and ‘iraq, under “ a ” mandates, the first being mandated to i'rance, and the two others to great britain. asia minor in the narrower sense has still been left entirely inside the turkish frontiers by the treaty of lau- sanne. on the other hand, the abortive treaty of sevres pro- vided for the partitioning of the narrower area also. a district round smyrna was to be placed under greek administration, while the remainder of the southern half of the region was divided between a french and an italian zone of special influence, under an anglo-franco-italian convention, signed simultaneously with the treaty. owing to the decisive turkish victory in the greco- turkish war of 1919-22, all these arrangements fell to the ground, so that the present area of turkey in asia practically coincides with asia minor in the narrower sense. (see ‘iraq; lausanne, treaty of; mosul; palestine; sevres treaty of; syria; turkey.) (age 2) 241 asquith, herbert henry: see oxford and asquith, earl of. | asthma: see respiratory system, diseases of the. aston, francis william (1877- ), british scientist, was born at harborne, birmingham, sept. 1 1877, and educated at malvern college and the universities of birmingham and cambridge. he was clected to a fellowship at trinity college, cambridge, and was made assistant lecturer in physics at bir- mingham university in r909. in 1920 he received the mackenzie davidson medal of the r6ntgen society, and in 1922 the hughes medal of the royal society. he received the nobel prize for chemistry in 1922 for his work in connection with isotopes and also the john scott medal, philadelphia, and the paterno medal, rome, both in 1923. he has written fsofopes (1922) and numer- ous papers in scientific periodicals on electric discharge in gases, muass-spectra, isotopes, etc. astor, john jacob (1886- ), younger son of the 1st vis- count astor, was born may 20 1886. educated at eton and new college, oxford, in 1906 he joined the ist life guards, and, from 1911 to 1914, was a.d.c. to the viceroy of india. in 1916 major astor married lady violet mary, daughter of the 4th earl of minto and widow of lord charles mercer-nairne. on the outbreak of the world war he went to france with the household cavalry and served there four years. he was wounded at the first battle of ypres and again severely in sept. 1918. in 1922, major astor purchased the holding of the late viscount northcliffe in the times publishing co., and became chair- man of it and chief proprietor of the times. shortly afterwards he caused to be established a body of trustees consisting of the holders of certain important non-political offices, whose consent he made necessary to any future transfer of the cantrol of that journal. he was elected conservative member for the dover division of kent in nov. 1922 and re-elected in 1923 and 1925. in 1925, as treasurer of the empire press union, he attended the press conference in australia. astor, nancy witcher, viscounrtess (1879- ), daugh- ter of chiswell dabney langhorne, was born may 19 1879 in virginia. she married in 1897 robert gould shaw of boston, from whom she obtained a divorce in 1903, and in 1906 she mar- ried waldorf astor. when her husband succeeded to the vis- countcy, lady astor, who had shown much interest in the affairs of his constituency at plymouth, was adopted as con- servative candidate, and was elected by a substantial majority on nov. 28 1919, thus becoming the first woman to sit in the british louse of commons.! she was re-elected at the general elections of 1922, 1923 and 1924. she spoke frequently in debate and was specially active in the cause of temperance, bring- ing forward a bill to prohibit the serving of liquor to young persons under 18 except at meal-times, which became law in 1923. , astor, waldorf astor, 2nd viscount, british politician, was born in new york may 19 1879, and educated at eton and new college, oxford. he represented plymouth as a unionist rg10-8, and the sutton division of plymouth 1918-9, when he vacated his seat on succeeding to his father’s peerage. he was chairman of the government committee on tuberculosis and of the state medical research committee. during the world war, he was inspector of quartermaster-general services, and in 1918 became private secretary to the prime minister, mr. lloyd george. he acted as parliamentary secretary to the ministry of food, 1918-9, and to the local government board in 19109, retain- ing the same position on the formation of the ministry of health up to rg2r. since 1915 he has been the proprietor of the observer and active on behalf of many causes of social progress, especially temperance reform. at the same time he was one of the leading british owners and breeders of race horses. his father, witttam walporr astor (see 2.794), became a naturalised british subject, and was created a peer in 1916, as- suming the title of baron astor of hever castle. he was made a viscount in 1917, and died in england oct. 18 1919. 1{n 1918 countess markieviecz had been elected by an irish constituency in the sinn fein interest, but she did not take her seat. 242 astronomy! (sce 2.800).—in the present century the centre of interest in astronomy has shifted from the solar system to the vaster domain of the stars and nebulae, and most of the progress in the ycars 1910-26 to be recorded relates to the stars. knowl- edge of the sun as 4 particular star has also advanced. the article star (see 25.784) was written ata time when statisti- cal methods of research were in their zenith. accumulated data as to proper motions, magnitudes, types, etc., for large numbers of stars were analysed and conclusions drawn as to the scale and structure of the stellar universe. after some further years of fruitful activity, this kind of stucly seems to have given place to more intensive study of individual stars. a feature of recent work is the application of modern atomic physics and the quan- tum theory of radiation to stellar problems; and the study of the stars has become closely linked with the study of the atom. it is now realised that the light despatched to us by the stars and nebulae contains claborate messages in a code to which atomic physics provides the key, and not only the condition of the radi- ating layers but the internal mechanism of the star can be in- ferred with some confidence. we divide the article into (1) the solar system; (2) the stars; (3) clusters and nebulae. i. the solar system the sun (see 26.85) by means of the spectroheliograph it is possible to obtain photographs of the sun in light of a single wave-length; we thus obtain a picture of the distribution of the matter which emits this wave-length or a negative of the matter which absorbs it. in practice either calcium or hydrogen light is used, since these elements furnish spectral lines sufficiently isolated to give good results. the emission of a particular spec- tral line depends on favourable conditions of temperature and density, and these will vary with the level in the sun’s atmos- phere. thus the function of the spectroheliograph is not so much to separate particular elements as to isolate different icvels in the sun’s atmosphere and provide separate photographs of what is occurring at each level. the highest level is given by photographs in the red light of hydrogen ha, and examples of the remarkable structure revealed by the spectroheliograph at the mt. wilson observatory are shown in the plate (figs. 5 and 7). in fig. 7 we see clearly the vortices whirling round the dark sun spots which occupy their troughs, and we can make out that the direction of whirl is opposite in the large spot and in the smaller spot to the left of it. if the whirling matter ts electrically charged it should act like a solenoid and produce a ficld of magnetic force; and this idea led g. e. hale (‘) to test whether a magnetic field could be de- tected in sun spots. by the zeeman effect a spectral line ab- sorbed or emitted in a magnetic field is broken up into two or more components; in particular for light travelling along the axis of the solenoid the line becomes a doublet with its two com- ponents circularly polarised in opposite directions. applying the optical test for circular polarisation clear evidence of the magnetic field in sun spots was obtained, the strength being gen- erally of the order 2,000 to 3,000 gausses. a large proportion of the sun-spot lines are observed to be slightly broadened, and the zeeman effect is at least partly responsible for this. it was pointed out many years ago by carrington that sun spots very frequently occur in pairs, the line joining them being approximately parallel to the sun’s equator. in these pairs the two spots are found to have opposite polarity, that is to say, if the magnetic field is upwards in the leading spot it is downwards in the following spot. even when the spot group 1s more complex a similar bipolarity is observed; hale estimates that in 90% of the spot groups the disturbed area shows this bipolar structure. the polarity follows a regular law. with rare exceptions the preceding spots in the northern hemisphere all have magnetic fields of one sign, and those in the southern hemi- sphere all have the opposite sign; after a sun spot minimum the signs in the two hemispheres are interchanged. this last reversal has been observed through two minima in 1912 and 1923 (*). 1the figures in parentheses appearing throughout the text of this article, refer to the references at the end. ; astronomy strangely enough there is no regular relation between the polarity of the spot and the direction of the whirl above it, and the vortices shown in the hydrogen photographs obey no such uniform laws as regards northern and southern hemispheres. this leads us to think that the origin of the sun spot and its magnetic field is down below. the spot may be caused by a vortex—formed by currents circulating in the interior of the sun——which rises to the surface and has a free end there. it is suggested by bjerknes that this internal vortex joins the follow- ing spot of one group to the leader of the next group, so that a spot group corresponds to a break in a permanent vortex in the form of a circle round the sun’s axis. this vortex travels from high to low latitude during the sun-spot cycle, and the sun spots follow it (as observation shows); it then descends into the inte- rior of the sun and at the same time a second permanent vortex of opposite rotation rises to the surface in high latitude and a new sun-spot cycle begins. the theory, though speculative and crudely developed at present, seems to have much to commend it. in any case it would seem that we have to do with a lower vortex —a whirlpoo/—formed according to uniform laws, and an upper vortex—a whirlwind—sucked in by it. hale has also found a general magnetic ficld of the sun, 2.e., not associated with exceptionally disturbed regions such as sun spots, which is roughly equivalent to the field of a uniformly magnetised sphere and analogous to the earth’s magnetism ('). it is found that the magnetic axis of the sun deviates from the rotation axis, though not so widely as happens on the earth, the inclination being 6 degrees. the synodic period of rotation of the magnetic axis is 31-44 days. ifwecould assume that the source of this field is the permanent magnetisation of the interior material this would give the real period of rotation of the sun—a quantity otherwise unknown. qour knowledge of the sun’s rotation has hitherto been based entirely on surface markings which revolve at different rates according to their latitude. the magnetic period of 31-5 days agrees with that of surface markings in latitude 55 degrees. the strength of this general magnetic field is of order 20 gauss and the whole observation is one of exceptional diff- culty, so that perhaps the foregoing determinations of the axis and period should not be accepted too hastily. the accepted value of the constant of solar radiation is that determined by c. g. abbot, viz.: the amount of solar energy pass- ing across 1 sq. cm. outside the earth’s atmosphere is 1-93 gram- calories per minute. this is the same as if the sun were a “ black body ” with a temperature 5,740° c. (absolute), which is accord- ingly defined to be the effective temperature of the photosphere. the sun’s total radiation is 3-8x10* ergs per second. there appears to be satisfactory evidence that the sun’s radiation varies a little according to the stage of the sun-spot cycle, being least when the sun’s surface is quiescent. abbot also claims that irregular changes varying according to the state of the surface have been established. since the same changes were meas- ured simultancously at two widely separated stations, terrestrial causes seemed to be ruled out (4). his earlier estimates of these irregular variations were undoubtedly excessive, and according to the more recent results they are restricted to very small range. another method of testing the variability of the sun has been carried out by guthnick and prager by photoelectric measure- ments of the brightness of saturn; the planet, being illuminated by ihesun, would reflect any changes in the intensity of the sun’s light. the measures showed that the light is nearly steady, and excluded the large fluctuations originally suspected. looking at the prob- lem theorctically, there is no special difficulty in admitting a small regular change in the rate of radiation according to the phase of the sun-spot cycle, but an irregular holding up and re- lease of considerable quantities of radiant energy seems unlikely. the photosphere is the name given to the region from which most of the sun’s light comes. although no definite limits can be assigned to it, its thickness may be taken to be of the order roo km. or perhaps less. the whole of this region is rarefied gas of density about jas of that of our own atmosphere; but the cocfficient of viscosity in these conditions is probably high, and hydrodynamically it may be compared to a slightly sticky astronomy fic. 1. monochromatic images of ring nebula. . &, fig 4 globular ¢ luster a, 3 (can . 3 prominence (may 29 1919). fic. 7. dunspot vortices astronomy liquid. the temperature ranges from not less than 4,650°c. at the top to about 7,000°c. at the bottom. the upper part of this region is called the reversing layer; this is where the great majority of the fraunhofer lines of the sun’s spectrum are imprinted on the light. the pressure here is about o-0001 atmospheres. above this rises the chromosphere, ex- tending to a height of over 10,000 km., which, it is now believed, is supported by radiation-pressure. general radiation-pressure is not very powerful on the sun, and the chromosphere is com- posed of a few elements (notably calclum) which by reason of strong line-absorption in the most intense part of the solar spectrum gain more momentum than the others. a detailed theory of the equilibrium of the calcium chromosphere due to e. a. milne (°) has proved remarkably successful. not only does it account satisfactorily for the height of the distribution, but it gives a purely astrophysical determination of an atomic con- stant of calcium—the life of an excited ca+ atom—which there is every reason to regard as correct. by special erup- tions matter can be carried to much greater height; fig. 6 (on plate) shows a prominence 100,000 m. high photographed by the british eclipse expedition at principe on may 29 igto. mars (see 17.761).—we shall review the problem of the physical and climatic conditions of this planet in the light of recent inves- tigations, including those made at the favourable opposition of 1924 (®). few astronomers have the opportunity of examining for themselves the phenomena reported to be observed at spect- ally favourable stations, since mars is never well placed for ob- servation in england and similar northern latitudes. conse- quently there is much doubt as to what should or should not be accepted as trustworthy; but recent great advances in the photog- raphy of the planet, especially at the lowell observatory, have removed scepticism on some points. the geometrical network of narrow rectilinear “‘ canals ” supposed by lowell and some other visual observers to cover the surface is still regarded with suspicion; most authorities explain it away as a mistaken impres- sion of faint markings—of curious form perhaps, but not dis- tinctively artificial. on the other hand some of the broader canals are shown on the photographs; and although the appear- ance does not particularly suggest artificial regularity, they are unlike any natural features on our own planet. seasonal changes of the surface markings are now conclusively established; a regular cycle of change with the martian season is repeated at each opposition. the dark markings appear in regu- lar succession, deepen in intensity and increase in area as the spring and summer advance, the polar snow melting meanwhile. it is difficult to resist the impression that we are viewing the annual growth of areas of vegetation on red desert soil. it is rash to accept so speculative a conclusion without more direct confirmation, but there seems to be no alternative explanation of the seasonal changes that is equally plausible. if life of any kind is admitted, can we restrict it to the vegetable kingdom? but that line of thought leads to guesses for which the astrono- mer accepts no responsibility. there is now strong evidence that mars has an efficient atmos- phere, though it is considerably more rarefied than our own. photographs occasionally show clouds which blot out the sur- face features for a time and pass away. there is usually a white fog over the polar region turned away from the sun, very different in appearance from the sharply outlined white snow cap covering the summer pole. a general haziness of the martian atmosphere is evidenced in a manner pointed out by w. h. wright. a terres- trial jandscape can be photographed by infra-red light through a mist so thick that nothing can be seen visually; the longer the wave-length, the less is the scattering and consequent blurring of the picture. photographs of mars in light of different wave- lengths behave just as though the surface-detail were being seen through a hazy atmosphere; ordinary photographic light of short wave-length gives little more than a general blur, photo- graphs in yellow visual light show more detail and those in infra-red light are the sharpest of all. determinations of the surface temperature of mars by radio- metric measurement of the heat coming to us from various parts 243 of the surface have been made by coblentz and lampland and by pettit and nicholson. the former obtained an average tem- perature of -+-15° c. and the latter—13° c. (other estimates arrived at on various assumptions are also given by the authors, but we prefer the foregoing based on stefan’s law.) evidently we are still left in considerable uncertainty as to the climatic conditions, but even the lower estimate is not inordinately cold when we remember that the polar snow cap is included in the average; the higher estimate makes the day temperature quite genial. according to coblentz the dark regions (indicating vegetation or moisture?) are warmer than the red desert regions. other planets (see 25.357).—a ninth satellite of jupiter was dis- covered by s. b. nicholson at the lick observatory in 1914. like the eighth satellite it revolves round the planet in the op- posite direction to the other seven. ‘the periods of satellites 8 and o are about 739 and 745 days respectively, and the two bodies are revolving in almost equal interlocked orbits in planes inclined at about 10 degrees. satellites 6 and 7 form a somewhat similar interlocked pair, their periods being 251 and 260 days respectively; but their motions are in the “‘ direct ” sense. the photograph of jupiter (taken at the lowell observa- tory) shown on the plate (fig. 2), illustrates the great advance in planetary photography in recent years. much interest has been taken in the trojan group of minor planets. these illustrate a special case of the problem of three bodies discussed by lagrange, viz.: that in which the three bodies are situated at the vertices of an equilateral triangle. ‘the tro- jan planets have the same mean distance and revolution period as jupiter, and the equilateral condition is roughly fulhlled. the problem of the small librations of such a planet about the triangular point of equilibrium has been discussed by e. w. brown; the condition of stability is that the mass of jupiter must be less than -0385 times that of the sun—a condition which is easily satished—and the period of the libration is about r4o years. actually the trojan planets are at some considerable dis- tance from the triangular points, and the problem of determining the finite librations (as opposed to infinitely small librations) has provided much exercise for mathematicians (7). six members of the group are now known, nos. 588 achilles, 617 patroclus, 624 hector, 659 nestor, 884 priam and 911 agamemnon; of these patroclus and priam are near the triangular point 60° behind jupiter, and the others 60° ahead of jupiter. the period of rotation of uranus round its axis has been de- termined by v. m. slipher from measures of the line of sight velocity of the advancing and receding limbs. the result is ro'sotm and the direction of rotation agrees with that of revolu- tion of the satellites. leon campbell subsequently found that the light of the planet is variable with the same period, presumably owing to unequal brightness of different parts of the surface. the rotation period of venus still remains a mystery; and there are advocates of the long period of 224 days as well as various estimates of a short period (one to three days). latitude variation (see 16.267).—the study of the small peri- odic motion of the earth’s axis of rotation (relatively to the earth) which gives rise to “ variation of latitude ” has been continued at the six international stations (reduced in number during and after the world war). the effect is made up of (a) the free pre- cession of a spheroid rotating about an axis which does not coin- cide with its axis of figure; the period of this precession deter- mined from the observations is 432} days; (0) an annual term, which is a forced oscillation due to meteorological and seasonal causes. owing to interference of these two terms, there is an effect analogous to “ beats” in sound waves, the amplitude of the motion alternately rising to a maximum of about 0”-3 (30 ft.) and dying out in about six years’ period. the annual term ap- peats to be nearly circular and of amplitude o”-o85; the possible causes contributory to this, such as seasonal circulation of the atmosphere and ocean, snowfall and vegetation have been in- vestigated by h. jeffreys, who finds a fair agreement between predicted and observed values. a mysterious kimura or z term, which appears in these international results, would, if interpreted literally, indicate an annual approach to the pole and recession 244 by all stations on the same latitude simultaneously—or a shifting of the earth’s centre of gravity to and fro along its axis. it is, however, now believed that the term arises from a small system- atic error in the observations; independent observations made at greenwich and pulkovo (not belonging to the international chain) show either a reduced or zero kimura term. gravitation.—the epoch-making theory of gravitation, put forward by einstein in rors, is described in the article relativ- ity (g.v.). we refer to it here because the new law of gravitation, required by his theory, removes the most outstanding divergence between theory and observation in the solar system—viz., the progression of the perihelion of mercury. there is still some dis- crepancy between theory and observation for the motion of the node of venus; but this is a much smaller residual, and may per- haps even be attributable to accidental errors. ejinstein’s pre- dicted deflection of light by the sun’s gravitational field was verified by the british eclipse expeditions in 1919 and further confirmed by the lick expedition in 1922. his third crucial test— a general displacement of spectral lines to the red in the sun as compared with terrestrial sources—was for some years a subject of controversy, as it is difficult to eliminate other causes of shift of the lines; but it now seems to be generally agreed among in- vestigators that the effect is confirmed. e. w. brown’s lunar theory, developed according to the meth- ods of g. w. hill, was completed by the publication in 1920 of full tables of the moon’s motion. it seems safe to say that no term of appreciable significance has been omitted; nevertheless the moon deviates unmistakably from its theoretical place in an irregular manner. an investigation by hf. glauert (*) seems to show that the irregularities are at least partly due to variations in the rate of our standard timekeeper, viz., the earth’s rotation; for the longitudes of the sun, mercury and venus exhibit similar irregularities, and the curves closely resemble one another. besides these irregular changes, there is a general secular acceler- ation of the moon, which, being cumulative, leads to large changes in the circumstances of ancient eclipses. the historical evi- dence of all kinds has been rediscussed by j. k. fotheringham (°) who arrives finally at the values 10”-5 for the moon’s secular acceleration! and 1”-o0 for the sun’s secular acceleration. these quantities are presumably attributable to tidal friction, which causes a direct acceleration of the moon’s orbital motion, as well as a spurious acceleration through the increase in the length of the standard of time. it is now believed that the bodily tides in the earth have little effect and that the most effective retardation is due to tides in land-locked and shallow seas. according to g.i. taylor the irish sea alone contributes 1/59 of the total dissipation of energy. ii. the stars progress in our knowledge must depend largely on the patient accumulation of accurate statistics as to the parallaxes, motions, magnitudes, spectra, etc., of large numbers of stars; we therefore review the great advance in these data that has taken place in recent years. the first photographic determinations of stellar parallaxes reaching a modern standard of accuracy were made by h.n. russell and a. r. hinks at cambridge and f. schlesinger at yerkes in 1903-7; earlier results are now superseded except for a few of the best heliometer measures made chiefly by gill. extensive programmes have since been carried out with large telescopes, and the latest compilation (1924) by the parallax committee of the international astronomical union contains results for about 1,870 stars. although the probable error is generally below + 0”-o1, this unfortunately does not give us the distances of 1,870 stars, for many of the parallaxes are inappreciable. very few of the stars are near enough for the trigonometrical method to give a close measure of their distance, and a large proportion of the measures can only be of interest in statistical discussions of distribution. we cannot resist the 1 the moon goes ahead by the amount 10”-5t?, where t is the time in centuries. this is the conventional definition of ‘‘secular acceleration ’’ in this connection. astronomy impression that stellar parallax determination is reaching its limit with present instruments; and perhaps for that reason special interest is attached to a new method of finding the dis- tances of stars described below under ‘ spectroscopic paral- laxes.”’ lewis boss’s preliminary general catalogue of 6,188 stars published in r91o has been an invaluable aid to research with regard to proper motions. it comprises all the brighter stars and the proper motions constitute a great improvement on data pre- viously available. the first really extensive lists of radial veloci- ties were published by the lick observatory in rg11. over 2,000 had been determined by 1921 when a catalogue was formed by j. votite. progress would have been more rapid but for the large proportion of spectroscopic binaries which makes it necessary to repeat the measures at suitable intervals in order to discrimi- nate between orbital motion and the steady secular motion which is looked for. the latest compilation (#9) records 1,054 spectro- scopic binaries, of which 248 have well-determined orbits. the apparent magnitudes of stars range from —1tm-5 for sirius to 20tm and upwards for stars photographed with the largest telescopes. the corresponding hght ratio is more than 100,000,- ooo to 1, and it is no easy matter to subdivide this range accu- rately. for this purpose a sequence of standard stars has been chosen near the north pole; their magnitudes stretch at short intervals from the first to the twenty-first magnitudes, and by. the co-operative effort of a number of observatories definitive magnitudes on an absolute scale have been found (#'). it is usual to determine magnitudes of other stars by differential com- parison with these. separate standards are needed for visual and photographic magnitudes; the relation of the two scales has been fixed by international convention so that they coincide for stars of spectral type ao between 5tm-5 and 6tm-5. since the photographic plate is most sensitive to blue light and the eye to yellow light, the difference photographic minus visual magni- tude gives a measure of the colour of the star. this is called the ** colour-index.”” the colour-index ranges from about —otm:5 for the bluest (type b) stars to -+1tm-9 for the reddest (type m) stars. the sequence of colour-index corresponds closely to the sequence of temperature, and the spectral type can usually be inferred correctly from the colour-index or vice versu. this affords a useful method of classifying stars too faint for spectro- scopic examination. some progress has been made with direct measurement of the heat received from the stars and the cor- responding ‘‘ bolometric”? magnitudes have been determined; considerable corrections are necessary on account of selective absorption of heat in passing through our atmosphere, and the method has been chiefly confined to the study of special stars. the bolometric magnitude is more usually inferred from the visual magnitude by applying a theoretical correction depending on the effective temperature (or spectral type) of the star. the draper classification of stellar spectra is now employed almost exclusively. the stages from the hottest to the coolest are denoted by the letters o, b, a, f, g, k, m; intermediate stages are estimated in tenths, e.g., g5 means half-way between go and ko. typical stars are b, rigel; a, sirius; f, procyon; g, the sun; k, arcturus; m, antares. types n and s seem to contain low temperature stars in conditions alternative to m, type m being characterised especially by the spectrum of tita- nium oxide, s by zirconium oxide and n by carbon compounds. a catalogue of the spectral types of 230,000 stars classified by miss a. j. cannon has been published by the harvard observa- tory and forms a monumental work of reference. the absolute magnitudes of stars cover nearly as wide a range as the apparent magnitudes. the sun’s absolute magnitude is 4-9, that is to say, it would have this magnitude if seen from the standard distance of 10 parsecs (parallax o”-1). the most lumi- nous stars have absolute magnitude about —5tm or 10,000 times brighter than the sun. the following are the four faintest stars known with their absolute visual magnitudes:—proxima cen- tauri 15tm-4, barnard star 13tm-3, groombridge 34 comes 13tm, pi 24 123 comcs 12tm-3. these are all red stars possessing a low sur- face temperature. astronomy masses of stars——in a double star the two components revolve round one another under their mutual attraction; it is possible to measure the gravitational force exerted by one component on the other and so determine the mass. there are not very many systems for which our knowledge is sufficiently accurate to give trustworthy masses; among the best determinations are sirius 2-45 and 0-8, capella 4-2 and 3:3, a centauri 1:1 and 1:o. the mass of the sun, viz.: 1-98. 10% gm., is here taken as unit. in spectroscopic binaries with known orbits of both components we can usually determine only a lower limit to the masses, be- cause one element of the orbit, namely its inclination to the line of sight, 1s lacking. these give evidence that very much larger masses occasionally occur. it has been shown by j. s. plaskett that the o-type star 6° 1,309 has components with masses not less than 86 and 72 times the sun’s mass. masses not less than 260 and 50 are attributed to the components of v sagitarii, but the evidence is less certain. v puppis consists of two equal stars of mass 19; in this case the star is an eclipsing variable, and the phenomenon of eclipsis tells us that the line of sight must be nearly in the plane of the orbit. at the other end of the scale masses from 14 to 4 the sun’s mass are common; the smallest known mass is that of the fainter component of krueger 60, which is probably about one-sixth. according to the theory of stellar radiation to be described presently, the absolute brightness of a star affords a close indi- cation of its mass. this theory is fairly well checked by the direct determinations of mass; and, if it is accepted, we can apply it to find the masses of the numerous stars for which no other determination exists. it appears that the stars are for the most part closely similar in mass; a range from 4 to 3 would include at least 90% of the stars. the larger meee mentioned above are quite exceptional; we tend to discover exceptional rather than normal masses because it is especially the brightest stars which attract attention. advantage is taken of this uniformity of stellar mass to deter- mine the ‘‘ dynamical parallaxes ” of double stars. if a is the semi-axis of the orbit in astronomical units, p the period in years, and mi-+mz the mass of the system in terms of the sun, we have by the law of gravitation my+ ig a?/p? thus a can be found if mj+m,z is known or guessed. we may assume with fair probability that m;-+mz= 2, the probable devi- ation being comparatively unimportant because a is proportional to the cube root of mi+my. but the value of ¢ in angular measure is found from the apparent orbit in the sky; comparing the angular measure with the linear measure given by the above calculation, we at once find the distance or parallax of the star. it is possible to modify the procedure so that it can be used when a small arc of the orbit has been observed insufficient for a complete determination of the orbital elements. dynamical parallaxes of 556 double stars have been published by j. jackson and h. h. furner (); from these the absolute magnitudes and linear velocities (transverse to the line of sight) were calculated. the linear velocities were combined to give a determination of the sun’s motion through the stellar system, the result being a velocity of 19:1 km. per sec. towards the apex r.a. 273°, dec.+34 degrees. this agrees remarkably well with the values generally accepted; and in particular the accordance of the speed with the value 19-5 km. per sec., obtained from the discussion of spectroscopic radial velocities, shows that the assumed mass 2-0 x sun must be almost exactly the average mass of a double star system. densities of stars.—the sun has an average density greater than water, viz.: 1-41. this 1s fairly typical of a majority of the stars; but in recent years we have learnt that there exist numer- ous stars of a much lower density. we have to enlarge our ideas to include “ giant ” balloon-like objects with density so rarefied as to be comparable with air or even with a moderately high vacuum. these giant stars are especially conspicuous because, being swollen to great bulk, they have great light-power. ex- cept for this high luminosity there is nothing very obvious about the appearance of the rarefied stars to distinguish them; the 245 ordinary spectral classification is no guide. for instance the three eclipsing variables w crucis, rz ophiuchi, sx cassiopeiae have densities less than -oor, but they are classed in type g along with the sun. it is now known that there are certain telltale fea- tures in the spectrum which can be used to distinguish them; but these are only found by minute inspection and were not noticed until after the giant stars had been found and studied by other methods. some information as to stellar densities can be obtained from eclipsing variables; it happens that a study of the eclipses gives sufficient data as to the relative dimensions of the two components and their orbit to solve the problem. but the most useful and universal clue is the absolute magnitude. the three stars above mentioned show practically the same spec- trum as the sun, and it seems fair therefore to conclude that their surfaces are in the same radiating condition and will give the same amount of light per unit area within reasonable approxi- mation. thus if a star of the same spectral type as the sun gives 100 times as much light (5 magnitudes brighter) its surface must be 100 times greater; its radius is accordingly 10 times greater and its volume 1,000 times greater. the inference can be ex- tended to other spectral types, accepting the principle that if the quality of the radiation as shown by its spectral distribution corresponds to a particular temperature the quantity of radiation must also correspond to that temperature at least to a rough approximation. whenever we know the absolute magnitude and spectral type of a star we can give a good estimate of its size. until recently this was our only means of determining the size of the stars (except for some meagre information from eclipsing variables), because the stars are so distant that no appreciable disk can be seen even in the largest telescopes. but in dec. 1920 the interferometer method developed by michelson was successfully applied to the problem; although the star disk could not be seen, yet its angular diameter could be measured by ob- serving the disappearance of a system of interference fringes (14). pease and anderson working with a 20-ft. interferometer at mt. wilson observatory measured the angular diameter of betelgeuse, the result being 0”:045. the parallax of this star is too small to be determined with much accuracy, and the con- version from angular to linear measure is a little uncertain; but it turns out that betelgeuse is certainly large enough for the whole of the earth’s orbit to be fitted inside it. a few other stars have since been measured, but most star disks are far too small to be within reach even of this powerful method. whilst it is valuable to have a proof by actual measurement that some stars have diameters 200 times greater than the sun, the importance of these interferometer results is that they confirm closely the diameters already deduced from the absolute magnitude and spectral type. as already exaleinea there is not a great deal of range in the masses of the stars, and the huge bulk of betelgeuse and similar stars indicates low density rather than great quantity of material. we have, of course, no actual measurement of the mass of betel- geuse, but we infer from analogy with other stars or from the theory of stellar radiation that it is not more than so times the mass of the sun. in any case it is certain that its density must be very low, because einstein’s theory shows that a star of the bulk of betelgeuse and density comparable to that of the sun would have extraordinary properties; in particular there would be a great shift to the red of the whole spectrum. according to the usual view the stars are formed by aggregation of primordial nebula, and it is therefore not surprising that we should find some stars in an early stage of low density. giant and dwarf stars—arranged in order of temperature from the hottest to the coolest, the spectral types fall into the sequence o,b,a,f,g,k,m. until about 1913 it was generallv assumed that this represented the course of evolution of a star, which started (presumably catastrophically) at high tempera- ture and gradually cooled. we have, however, just seen that betelgeuse and some similar stars which are cool red stars of type m are in an extremely diffuse state, and therefore represent the earliest stage in the condensation of a star from a nebula, we must not jump to the conclusion that all type m stars are 246 to be placed in the earliest stage of evolution. closer examina- tion shows that type m is heterogeneous and contains two quite distinct groups, the one consisting of giant diffuse objects and the other of small dense stars (red dwarfs) apparently in the last stage before extinction. this and other features of the spectra, luminosities and densities of the stars are brought into order by the giant and dwarf theory of evolution put forward independ- ently by e. hertzsprung and h. n. russell, which has dominated research in stellar astronomy from 1914 to 1924. there is now strong reason to think that the theory cannot be sustained in its present form; but since it undoubtedly contains a great deal of truth and the necessary reconstruction has not as yet made much progress we shall give the usual form of the theory and later point out where it fails. we must go back to homer lane’s theory of the evolution of gaseous masses (see 25.788). starting with a very diffuse globe of gas held together by its own gravitational contraction, the conditions of mechanical equilibrium require that its tempera- ture must rise as it contracts, this rise of temperature continues so long as the material is rare enough to behave as a perfect gas. the gases with which we are familiar cease to follow the laws of a perfect gas when the density approaches that of the liquid state; it may therefore be anticipated that in the sun (density 1-4) the material is behaving more like a liquid and the star is cooling in the ordinary way. accordingly the idea is that a star starts as a cool diffuse mass of perfect gas; it condenses and rises in temperature until the deviations from the gas laws check the rise of temperature; it then cools like a liquid and finally be- comes extinct. it follows that a star passes through any particu- lar temperature and spectral type twice, once ascending as a diffuse giant and once descending as a condensed dwarf. in any spectral class we shall have two groups of stars alike in surface temperature but wide apart in all other respects; in particular the two groups differ in density and in stage of evolution. the most conspicuous outward characteristic is the great difference of luminosity due to the larger surface area in the giant stage. instead of having a single evolutionary sequence of spectral types, the star must start as a giant of type m, ascend toward type b and then descend as a dwarf to type m again. it de- pends on its mass how far up the series it climbs, and probably a star must be three or four times as massive as the sun in order to reach the high temperature of tvpe b. smaller stars will turn at a, f or even lower. as russell has put it, a star of small mass is a poor self-heating affair. the division of giants and dwarfs is most conspicuous for the lowest temperatures k and ml since the corresponding stages are then at opposite ends of the evolu- tionary sequence; for types f and a the two groups begin to merge into one another and the division js less easy to recognise. the observational confirmation drawn from many sources is now extremely favourable. for stars of known parallax the absolute luminosity can be calculated directly; and when these are grouped according to spectral type the bifurcation of the luminosities is evident. the luminosities of the giant stars de- pend very little on the spectral type since the rising temperature compensates for the decreasing surface area; their absolute mag- nitudes cluster closely about a mean value +1tm-0. for the dwarfs the decreasing temperature and decreasing surface co-operate to give a rapid fall of brightness through successive types, and the absolute magnitude fades to about + 10-0 at type m. by the spectroscopic method of determining absolute brightness, adams and joy have been able to give striking statistics; of 58 stars of type m examined. they found that 48 were giants with magnitudes between—1\"-o and +3tm-4, and ro were dwarfs be- tween +9\"-8 and +10”-7; there was thus a gap of six magni- tudes separating the groups. (ut is possible that the sharpness of the separation is exaggerated by the circumstances of sclection of the stars for examination, but we do not think the influence of selection is so great as to make the result misleacling.) ascendl- ing to types k and g the groups draw closer together and begin to commingle, but even in type f the frequency curve shows the two maxima, the bifurcation into giants and dwarfs is also well shown in shapley’s investigation of the densities of eclipsing astronomy variables and in jackson and furner’s study of dynamical paral- laxes. | before leaving the observational evidence we must refer to a point which becomes especially significant now that russell and herizsprung’s theory of evolution is under suspicion. the giants apparently do not link on continuously to the dwarfs at type a. we trace the regular giant series from type m to g, and then there seems to be a gap with scarcely any normal giants of type f—as though this stage were rushed through rapidly. such giants as do occur in type f are of higher luminosity and mass. we are now more inclined to divide the stars into a “ main series’? running down in temperature continuously from in- tenscly luminous stars of type o to faint stars of type m, together with a side group of giants chiefly limited to types g, k and m. the latter is presumably a temporary halt in the early stages of evolution; when the halt is over the star proceeds rapidly to join the main series at a point depending on its mass. the shock to the giant and dwarf theory comes from a result reached in 1924. both from astronomical observation and purely physical theory it seems that stellar matter will continue to be- have as a perfect gas up to densities exceeding that of platinum. thus there is no explanation of the turning point from the giant to the dwarf series, and the descent along the dwarf series (if it occurs at all) must be due to some entirely different cause. radiative equilibrium of the stars —the discovery that many of the stars are of very low density has given a stimulus to in- vestigations of their internal conditions of equilibrium; for the material, being practically a perfect gas, will obey comparatively simple and definite laws. the distribution of temperature and density in a sphere of gas in equilibrium under its own gravita- tional attraction is a classical problem studied by lane, ritter, kelvin, emden and others. for the most part the mathematical analysis developed in these earlier researches is used in the mod- ern theory, but three new features have been introduced :— - (a) it used to be supposed that the equilibrium was adiabatic— that is to say, the materia] was kept stirred by convection cur- rents, hot gases ascending to replace the continually cooling ma- terial at the surface. but it is now clear that the heat passes to the surface not by material transfer but by radiation, and the condition of equilibrium is that each region will settle down to the temperature at which it radiates an amount of heat equal to that which it absorbs from the radiant heat tlowing through it. this was first suggested as probable by r. a. sampson in 1894, and the theory of radiative equilibrium was developed by k. schwarzschild in 1906 for the atmosphere of the sun. it was taken by the writer as the basis of the theory of the stellar interior in rgt6. one simplification resulting from this change is that a physical constant of stellar material, viz., the ratio of specific heats, which was difficult to estimate, is no longer re- quired to be known. . (b) the pressure of radiation is sufficient to have an impor- tant effect on the condition of equilibrium especially in the more massive stars, and is taken into account in the new theory. | (c) it is recognised that at the high temperature in the stellar interior the atoms will be strongly jonised. most of the electrons which circulate round the nucleus of an atom will have broken loose; and these, moving freely and exerting their own partial pressure, must count as separate ‘‘ molecules ” in determining the average molecular weight. all numerical results depend on the adopted value of the molecular weight; and it is essential that this should be known rather accurately since it 1s raised to a high power in several of the most important formulac. owing to the ionisation we adopt a much smaller molecular weight than in the earlier investigations. moreover we are no longer scriously troubled by uncertainty as to the chemical composition of the stellar interior, because whaterer the chemical com posilionn (pro- vided only that there is not an excessive proporijon of hydrogen) the molecular weight will be close to two. this is a consequence of the fact that the number of electrons surrounding the nucleus of an atom is approximately half the atomic weight, so that when all have broken loose there will be one free particle for every two units of weight. in stellar-conditions the stripping of the atom astronomy is not quite complete, but the modification of the molecular weight can be calculated and allowed for. since it is believed that the star’s heat is maintained by libera- tion of subatomic energy, another unknown condition enters into the problem, viz.: the relative distribution of the subatomic source through the interior. this, however, has no very impor- tant effect on the results, and the margin of uncertainty due to our ignorance of the distribution has been calculated by consid- ering the extreme cases of a source entirely concentrated at the centre and a source fairly evenly distributed through the star. all the difficulties and uncertainties attending the calculation of the internal temperatures of the stars seem to have climinated themselves, and it is believed that we have trustworthy knowledge for all stars in the condition of a perfect gas. the effective temperatures of a few thousand degrees represent only the marginal heat of the stellar furnaces, and the greater part of the material of a star is at a temperature of several million degrees. in a typical giant star capella the central temperature is 10,000,000° and the average temperature is about 2 of this. in the sun the central temperature is about 40,000,000°. (we shall see later that the theory applies to the sun in spite of its high density.) all stars on the main series have nearly the same internal temperature as the sun. it should be realised that from the standpoint of the expert in atomic physics these tempera- tures are not at all high. temperature is a measure of the aver- age speed of the molecules. at 40,000,000° the speed of the free electrons is still low compared with that ordinarily imparted to them in electrical experiments, and far less than that of the 6 particles shot out in radioactive disintegration. we cannot antici- pate any disintegration of the nucleus, such as rutherford has found in the laboratory by employing incomparably more power- ful agency. the difference is that, whereas qualitatively the atomic processes in a star are by no means out of the ordinary, the intensity or rapid succession of the individual processes is not reproducible in laboratory experiments. it helps us a great deal that in studying the stellar interior we have not to envisage unfamiliar atomic processes but to calculate the cumulative effect of a great multiplication of familiar processes. the tempcrature-gradient in the interior tends to cause a flow of radiation outwards, which is hindered by the opacity of the stellar material, the opacity being a measure of the amount of obstruction offered by a substance to the flow of light or other radiation through it. thus the total outflow of heat from the star depends on the internal temperature distribution (calculated by the theory already discussed) and on the opacity. if l is the radiation of the star (ergs per sec.), we find l=a4mcegm(1-8)/k where c is the velocity of light, g the constant of gravitation, m the mass of the star (in grammes), & the coefficient of opacity, and 1-g the ratio of radiation pressure to the whole pressure found by solving the quartic equation | i-b=0:0031 m? py! £4 where m is now expressed in terms of the sun’s mass and yp is the average molecular weight. knowing il and m for any star we can now calculate &; for instance in capella we find k= 120c.g.s. units; this means that radiation passing through a screen of stel- lar material containing 1/120 grammes per sq. cm. (equivalent to about 6cm. of air) would be reduced to one-third (strictly 1/e) of its original intensity. this is a very high opacity. we see that the old difficulty as to how heat is brought up from the interior of a star to replace that radiated by the surface material has vanished; there is no need for convection currents; our problem is rather to explain how the material offers so much obstruction and dams back the internal heat, allowing it to escape no faster than it does. the high opacity is not so surprising when we realise that the radiation in the interior of a star consists of x- rays chiefly of wave-length 1 to ro angstrom units, and the opac- ity determined astronomically is of the same order as the opacity of ordinary material to these x-rays as measured in the labora- tory. the stellar opacity is in fact less than the terrestrial opacity because the high ionisation of the material throws a great part of the absorbing mechanism out of order. 247 the mass luminosity relation.—application of the modern physical theories of x-ray absorption to stellar conditions indi- cates that the coefficient of opacity will be directly proportional to the density and inversely proportional approximately to the ; power of the temperature (& 0¢ t?). by the use of this law we eliminate & and obtain a relation between land m. the density of the star nearly disappears from the relation; so that the total radiation or absolute bolometric magnitude of a star is a function of the mass, subject only to a trifling correction depending on its stage of evolution as indicated by density or by spectral type— provided always that the star is sufficiently diffuse to behave as a perfect gas. we call this predicted relation the mass-lumi- nosity relation. so far as it can be tested with the not too abundant observational data available it appears to be strongly confirmed. the physical theory has not been so successful in predicting the absolute scale of luminosity of the stars; at present the position is that whilst the stars agree among themselves, they are all apparently about 2} magnitudes too faint. the cause of this discrepancy remains a mysterv; but so long as we confine attention to differential comparisons it does not trouble us. | the startling result appears that not only do the giant stars con- firm the mass-luminosity relation but the dense dwarf stars also agree with it, and the sun, for example, falls into line on the mass-luminosity curve (4). taken at face value this unlooked- for agreement indicates that the material of the sun must be behaving as a perfect gas. this would be quite opposed to the giant and dwarf theory of evolution which attributes the low luminosity of the sun at its present stage (compared with that of the diffuse stars) to the deviation from a perfect gas. never- theless we seem bound to accept the new conclusion not only on account of the argument by which it was first reached, but be- cause consideration of the physical conditions of material at a temperature of some millons of degrees clearly shows that it will remain a nearly perfect gas up to extremely high densities. ter- restrial material ceases to behave as a perfect gas at densities approaching that of water, because the atoms, which behave as rigid spheres with radii of the order 1078 cm., are becoming jammed together. but in the stars atoms of this description do not exist; there are only the small fragments of atoms (ions) with radii of order 107! cm., and these do not jam until a density some thousands of times higher is reached. our expectation that there would be a change in stellar conditions at a density between say o-r and 1 rests on a false analogy, which proves to be baseless as soon as attention is called to it. detailed investigation, taking account of the electrical forces between the ions, shows that there will be no appreciable failure of the gas-laws in any ordinary stars—except that in stars of small mass the material will be swperperfect gas causing these stars to be about 4 a magnitude brighter than at first predicted. there is some slight observational confirmation of this excess. the theory becomes extended to include all ordinary stars (i.e., excluding ‘ white dwarfs ’’) and differences of absolute bright- ness are now attributed almost entirely to differences of mass. this fundamental change of view was only reached in 1924, and there has scarcely been time fully to reconstruct our ideas of stellar evolution in accordance with it. the companion of sirius.—if matter at the density of plati- num is still behaving as a perfect gas the limit of compressibility must be at far greater density. thus it is possible that stellar matter may exist having a high density transcending terrestrial experience. conversely if definite evidence of such matter can be found it will be the strongest possible confirmation of the views to which we have been led. it happens that there already existed evidence of the existence of stars of extravagantly high density—evidence which would probably have been accepted as fairly convincing if the conclusion had not been thought too absurd to be taken seriously. these stars are called w/ite dwarfs. only three are definitely known; but they are probably rather numerous in space because as a rule they are not detected unless thev are close to us. the most famous is the companion of sirius which has 3 the mass of the sun but gives only 1/360th of the 248 sun’s light. this would not surprise us if it were a red star of feeble surface luminosity, but it was shown by w. s. adams in r919 that it was a white star of spectrum f and therefore should have a surface-luminosity greater than that of the sun. the radius calculated in the ordinary way from the luminosity and type is 19,000 km.—not so large as uranus—and the correspond- ing mean density is 60,000 grammes per cubic centimetre. when it was realised that these results were not necessarily nonsensical, a further test was applied by means of einstein’s theory. the red-shift of the spectral lines—the third einstein effect— is proportional to the mass divided by the radius and according to the above figures would be 30 times as great on the companion of sirius as on the sun. it would be equivalent to a doppler effect of 20 km. persecond. (the true doppler effect can be sepa- rated from the einstcin effect because the motion in the line of sight is known from observation of sirius itself.) although the observation was difficult on account of the faintness of the object and its proximity to sirius, adams was able to show in 1925 that an einstein effect of 21 km. per sec. existed, agreeing with the prediction (5), the companion of sirius has been famous in the history of astronomy as a body whose existence was known and accepted unreservedly 18 years before it was first discovered with a telescope, bessel having shown in 1844 that sirius was revolv- ing under the attraction of an unseen mass; it now comes into prominence again as revealing that stellar matter can be com- pressed to a density of a ton to the cubic inch. there is no reason to think that this matter consists of other than the familiar ele- ments; it is simply a case of tighter packing of the smashed atoms. radiation pressure —at any point in a star gas-pressure and radiation-pressure together support the weight of the mate- rial above. the proportion supported by radiation-pressure (1-8) is to a first approximation the same in all parts of the star and, if the molecular weight is given, depends only on the mass. we might expect something significant in the range of mass for which radiation-pressure increases from minor to major impor- tance, say 1-8 increases from o-1 too-5. the masses correspond- ing to these limits are 1-6 to 11-5 times the sun’s mass (for mo- lecular weight 2-1). this seems to give a clue to the remarkable phenomenon that the masses of the stars are so nearly uniform. apparently the matter of the universe has aggregated primarily into lumps whose size is determined by this critical range of radiation-pressure. we know that a gaseous mass without radi- ation-pressure is stable unless it is in extremely rapid rotation; we do not know the conditions of stability of a mass subject to radiation-pressure. it is a plausible hypothesis that high radi- ation-pressure endangers the stability of a star;in that case the aggregation of matter to form stars will be put a stop to at limits of mass generally in the range indicated. in considering how accurate this coincidence is we must notice that there is now a wide-spread belief that the star radiates a considerable part of its mass during its lifetime. if so we must take for comparison the original masses, or the masses of stars observed in the most diffuse state. the range 1-6 to 11-5 agrees excellently with the known range of mass of the k and m giants. saha’s theory of stellar spectra.—after the first use of spec- troscopy to identify the chemical elements in the atmosphere of a star, the most important advance was the discrimination of “ enhanced lines,” i.e., lines strengthened in the spark spectrum as compared with the arc. this gives a criterion as to the physi- cal condition of the star’s reversing layer, because it depends on the temperature and density whether the arc or the spark spec- trum predominates. much of the earlier progress in this devel- opment is due to sir j. n. lockyer and a. fowler. in general the arc lines are absorbed by the neutral atom and the spark lines by the atom with one electron missing (singly ionised atoms). sometimes the distinction can be carried further; and for ex- ample, fowler has identified the separate spectra of singly, doubly and trebly ionised silicon in the laboratory and in the stars. the modern thermodynamical theory of ionisation was brought to bear on the astronomical observations by m. n. saha in 1920 and his method of interpretation of spectra has domi- astronomy nated all recent progress (16). consider the element calcium. ina red star the temperature of the reversing layer is low and the atoms will not be ionised, so that only the arc lines appear. at somewhat higher temperature ionisation begins, and the h and k lines of ionised calcium are shown. in the sun both spectra are prominent. passing to higher temperatures the arc lines disappear since no atoms are left in the neutral state; and in the hottest stars the atoms become doubly ionised so that the h and k spectrum fades out. if the pressure in the reversing layer is known we can calculate by theory how the state of ionisation progresses with tempera- ture and so determine the temperatures corresponding to the spectral types in which the various spectra are observed to ap- pear, disappear or reach maximum intensity. lines absorbed by exciied atoms (ionised or not) are especially useful, since the abundance of excitation is sensitive to small changes of tempera- ture. by connecting together the data for different elements and controlling the results by reference to the best determinations of stellar temperature by other methods, a fairly consistent tem- perature-scale for the various types has been found. it appears that the pressure in the reversing layer is between 107% and 1074 atmospheres—much lower than used to be supposed. a few lines are found to correspond to a still lower pressure; these are probably chromospheric lines absorbed by ions which experience strong selective radiation pressure which maintains them at a high level above the ordinary reversing layer. a number of the improvements in the detailed applications of the theory are due to r. h. fowler and e. a. milne. spectroscopic parallaxes——although giant and dwarf stars of the same temperature have, broadly speaking, the same spec- trum, a detailed examination of particular lines reveals distinc- tive differences. it was early shown by e. hertzsprung that those spectra marked by miss maury as having “ c-character- istic’ belonged exclusively to giant stars. more precise criteria were found by w. s. adams and a. kohlschiitter in 1914; and the method has been developed by adams into a means not only of distinguishing the two classes but of determining quanti- tatively the absolute luminosities of stars. for example, the ‘“‘ enhanced lines ’’ of strontium 4,077 and 4,215 are relatively strong in stars of high luminosity and weak in those of low lumi- nosity; whereas the “ furnace lines” of strontium 4,607 and calcium 4,455 behave in the reverse manner. thus measures of the relative intensities of these lines give an indication of the luminosity of the star. at present the procedure is purely empirical; the curve con- necting absolute magnitude with differential intensity of the selected lines is deduced from and tested by stars of known trig- onometrical parallax; it is then applied to deduce the luminosi- ties and hence the parallaxes of other stars. parallaxes deter- mined by this method for 1,650 stars have been announced ('’). the existence of this criterion of absolute magnitude is explained in a general way by saha’s theory. the state of the outer layers of a star, and hence the spectrum, is controlled by two factors, viz.: the value of gravity at the surface and the intensity of the stream of radiation poured through them from below; the latter factor is equivalent to the effective temperature. take two stars of high and low luminosity respectively, e.g., a giant and dwarf of the same effective temperature; the former being a dif- fuse star has a much lower value of gravity, and this has the consequence that the reversing layer occupies a level of lower pressure. with lower pressure the same stages of ionisation occur at reduced temperature; and it is in fact confirmed that the same spectral type corresponds to lower temperature in a giant than in a dwarf. but a change of pressure does not simply shift the temperature scale for the whole spectrum; actually the shift is different for each element and each phase of the atom— normal, ionised or excited—so that differential comparisons of particular lines betray the alteration of pressure. variable stars—the three leading classes of variable stars are (a) eclipsing variables, (b) cepheid variables, (c) long-period variables. rather more than 200 eclipsing variables are known; in these one component wholly or partly hides the other during astronomy part of the revolution so that there is a decrease of brightness. intensive study of the light-curves of particular stars gives im- portant information as to the radii of the components, densities, surface brightness, etc. in a number of systems a “ reflection effect ’’ has been observed and measured; that is to say the two hemispheres of the faint component are of unequal brightness, and the star goes through phases like the moon owing to illumi- nation by the bright component. it has been shown theoreti- cally that a star must be a perfect reflector of heat (heat—albedo = 1); but since our actual measurements are not in heat units but in light units this cannot be immediately checked by observa- tion. in some systems the two components are so close that they distort one another, and the prolate form of the stars gives rise to observable features in the light-curve. it would seem that eclipsing variables will afford a quite exceptional opportunity of studying details of stellar constitution, but progress at present is difficult. | the explanation of the cepheid variables is more uncertain. the first question is: is the cepheid a binary star? the spectro- scope apparently answers in the affirmative for it shows a radial velocity increasing and decreasing in the period of the light vari- ation, as though the star were revolving round an unseen body. but the change of light is not due to eclipses because the mini- mum brightness occurs when the star is receding most rapidly— at a time when the other component could not be between us and it. there must be an actual variation in the rate of radiation by the star, and this has been confirmed by h. shapley who showed that the spectral type and therefore the surface condi- tions change during the period. for example, 6 cephei changes from type fo at maximum to g2 at minimum. this periodic heating and cooling is the main cause of the change of brightness. one suggested explanation is that the orbital motion occurs in a resisting medium, so that the front side of the star is brighter than the rear side on account of the impact of the medium; this would explain why minimum brightness always occurs when the star is retreating. but opinion is now tending towards a pulsa- tory theory proposed by h. shapley () which rejects the binary hypothesis altogether. the fact is that there is literally no room for the supposed second component required by the binary hy- pothesis. the cepheids are giant stars filling a large volume, and the “ orbit ” is always small compared with the dimensions of the star itself. when we calculate the size of the orbit of the supposed companion (which we can do, knowing the period and approximate mass of the system) we find that it would graze or even lie inside the principal star—a reductio ad absurdum of the binary hypothesis. further, a relation has been found between period and density in these stars, which points to the period being determined by intrinsic conditions; such a relation is quite un- intelligible if the period is provoked by an external cause, viz.: the revolution of a companion. accordingly shapley suggests that the variable 1s a single star which dilaies and contracts with a vegular pulsation; and the observed motion of approach and re- cession refers, not to the star as a whole, but to the upheaval and subsidence of the part of the surface presented towards us. the radius of 6 cephei may be taken as about 20,000,000 km.; the semi-amplitude of the oscillation, according to the observed radial velocities is 1,270,000 km. or about 6% of the radius. for 15 other fully observed cepheids the semi-amplitude of the pulsation ranges from 3 to 8% of the radius; this seems an amount of compression and expansion suitable to produce the rather large changes of temperature required. within narrow limits the period is inversely proportional to the square root of the star’s mean density, a relation which seems significant in view of the fact that the pulsations of a gravitating sphere follow this law. moreover the constant of proportionality is of the order of magnitude predicted by theory: we can calculate that a globe of gas having the mass and density of 6 cephei will vibrate in a period between 4 and 10 days (varying between these limits according to the adiabatic constant of the material of which it is composed); the observed period is 5-37 days. (%) the most serious objection urged against the pulsation theory of cepheids is that it requires a broadening of the spectral lines at minimum 249 and maximum, because all parts of the disk would not be moving with the same speed in the line of sight; this has not yet been observed. it is to be hoped that this crucial but rather difficult effect will be thoroughly sought for in the near future. a question arises as to how the pulsations are maintained. there is very little dissipation of energy from viscosity and the main cause of decay of the pulsations would be the leakage of heat between different parts of the star. we have in any case to admit that a supply of energy (sub-atomic energy) is continually being released in the stars, and if certain thermodynamical con- ditions are satisfied this may supply the mechanical energy re- quired to maintain the pulsations. the simplest theory is that this energy is released more rapidly when the star is compressed and at maximum temperature; the star then receives excess heat at high temperature which it radiates at low temperature and so acts as the working substance of a heat-engine. this expla- nation is perhaps too powerful, as it makes it dificult to under- stand why every star is not set in pulsation. an alternative theory is that the heat supply is constant but the varying opacity (acting like the valves of an engine) governs its flow in such a way that mechanical work is performed. it now seems probable that the long-period variables link on continuously to the cepheid variables, the differences being only skin-deep. it is, however, the skin which we observe, and in the long-period variables, which are stars of very low density and low photospheric temperature, a great surface eruption occurs at a certain stage of the pulsation. at the height of the eruption the star is usually at least 100 times as bright as at minimum; but it has recently been discovered that the increase of heat radiation is not nearly so great. by direct radiometric measurements, the variation of heat is found to be about 1 magnitude—just as in the cepheids. the theoretical relation between period and den- sity obeyed by the cepheids is found to apply also to these low density stars, and the fact that the long-period variables are all of type m (or the other red types n and s) agrees with the rule found for the cepheids that the type becomes redder with in- creasing period. novae.— two “ new stars ”’ of unusual brilliance have appeared in recent years. nova aquilae iii. was discovered independently by a great many observers on june 8 1918 when it was already a first-magnitude star. its earlier history has been supplied from an examination of photographic records of the sky. from 1888 onwards it remained steady at rotm-5 and a photograph taken by max wolf three days before discovery showed that it was still normal, incidentally we may note that it cannot have been a red star (types k or m) or it would have appeared in visual cata- logues. on june 7 it had reached 6tm- according to a harvard photograph. the next day (when it was discovered) it had brightened to otm-8; and on june g it was only slightly inferior to sirius. then followed the usual slow decline with occasional fluctuations; and it had faded to 5tm-5 by theend of october. w.f. denning discovered a nova in cygnus on aug. 20 1920 which reached the second magnitude. its earlier history is unknown, but it must have been fainter than 15tm- in 1908. broadly speaking each nova reproduces the same sequence of phenomena with remarkable faithfulness. if caught early enough before maximum brightness is reached, the spectrum is that of an a or b type star with the absorption lines displaced to the violet. this appears to indicate that the outer layers of the star are being propelled upwards; the velocity varies from day to day. a few days later bright emission lines appear on the red side of the corresponding absorption lines. the absorption lines become doubled and trebled as though there were several jets of uprushing gas with different speeds. the bright lines broaden into bands and show complicated structure. the light is now declining owing to the continuous spectrum fading away, and the spectrum becomes mainly an emission spectrum. about a fortnight after maximum bright nebular lines (7.e., lines of un- identified origin but characteristic of gaseous nebulae) appear. the great speed of upward rush of the absorbing gases is very remarkable, velocities of the order 2,000 km. per sec. being ob- served; there is no reason to doubt that these velocities are 250 genuine for the star does actually expand and in the later stages shows a nebulous disk visible in large telescopes. the observed rate of spreading appears to agree with the speeds indicated by the shift of the absorption lines. after the outbreak has subsided the star becomes of spectral type o surrounded by a nebulous disk—like a planetary nebula. for this reason it is sometimes thought that the ordinary planetary nebulae originate from novae; but this seems unlikely, because, so far as can be ascer- tained, there is a big difference in scale of the two kinds of objects. of the many theories advanced to account for the outbreak, a collision of two stars seems unlikely on account of its statistical improbability; and moreover the regular sequence of changes could scarcely be started by a haphazard impact. an eruption from within, whether occurring spontaneously at a certain stage of evolution or precipitated by the entry of a star into a nebula, may be more likely; but this theory also presents difficulties. j. h. moore has shown that the extended nebulous disk shows differential motions of rotation in different parts. in any case, it seems likely from the very rapid sequence of changes that the main outbreak is only skin-deep. novae always occur within the limits of the milky way (or in spiral nebulae); but this may per- haps be due to the greater depth of the stellar universe in this direction. so far as can be judged the nova before the outbreak is a dwarf star; and at least in the case of nova aquilae it cannot have been a very red star. (the long-period variables, whose violent outbreaks are rather suggestive of the explosion of a nova, are giant red stars.) we may meditate on the fact that the stars subject to these catastrophes are probably in about the same stage of evolution as that through which the sun is now passing. | | motions of the stars ——~many researches have confirmed kap- teyn’s discovery that the stars (or at least those near enough for investigation) move preferentially in two favoured directions. since the article star (see 25.784) was written, the spectroscopic tadial velocities have become available for testing the theory and they confirm it decisively. relatively to the sun the favoured directions are inclined at about 120° (the apices being at r.a. 96° dec.+8°, and r.a. 290°, dec.—54°); but referred to the mean of the stars they are necessarily two opposite directions along a straight line. the extremities of this axis of preferential motion are called the vertices. the following appear to be the most accurate determinations of the vertex by the two indepen- dent methods (°) :— | from proper motions (boss’s cataloguc) r.a. 94°-2, dec. +11°-9, from radial velocities (lick catalogues) r.a. 94°-6, dec.-+12°°5. - we should rather expect this star streaming to be a local phe- nomenon in the stellar universe, but up to the present no sensible change in the velocities or relative abundance of the two streams has been found in the region surveyed. photographic determina- tions of the direction of motion of faint stars of small proper motion made a cambridge observatory show that the phenome- non occurs without detectable alteration up to distances of 500 parsecs from the sun. it is significant that the line of preferential motion lies in the galactic plane. the phenomenon may be due to two star clouds passing through one another; or it may repre- sent some dynamical property of a single system. the latter view has often been favoured owing to the very elegant mathe- matical specification of the corresponding velocity distribution given by k. schwarzschild’s ellipsoidal theory (7). the more detailed study of the systematic motions of the stars leads to results of great complexity. we have to recognise a third drift, pointed out by j. alm, which is nearly at rest rel- ative to the mean of the stars. the type b stars appear to belong to this third drift and show no sign of the preferential motion towards kapteyn’s vertices; but the drift also comprises stars of other types. another important phenomenon is the asymmetry of the motions of high velocity stars. we shall here follow the conclusions of j. oort (*) with regard to this phenomenon; g. stremberg, who has also investigated it, holds somewhat differ- ent views. according to oort stars with velocities above a cer- tain limit (probably about 60-80 km. per sec.) are moving al- most exclusively towards one hemisphere of the sky. this is an astronomy additional effect, apparently having nothing to do with the two main star-streams. tentatively we suppose that these high ve- locity stars are intruders into our local star cloud, having come into it from another cluster situated on one side. regarding our star cloud as a semi-permanent system, there will be a certain velocity of escape. stars with velocities below this limit are per- manent members of the system describing orbits under its attrac- tion, and there cannot well be one-sided asymmetry in the dis- tribution of such orbital motions; but stars with speeds greater than the velocity of escape must be intruders which pass through the system once only, and preserve more or less the direction of motion with which they entered. whatever the true explana- tion, these high velocity stars all moving one way affect consid- erably the various statistical discussions of the relation of stellar motions to other characteristics such as type and luminosity, because the asymmetric stream appears to include a larger pro- portion of stars of low luminosity and reddish type. in 1910 j. c. kapteyn and w. w. campbell announced inde- pendently that (after allowing for the solar motion) the average speeds of the stars increase continuously from type b to type m. according to campbell the average radial speeds are—type b, 6-5; a, 10-9; f, 14-4; g, 15-0; k, 16-8; m, 17-1 km. per second. in this investigation the k and m stars were almost all giants; but it has been found that the dwarf k and m stars have still higher speeds averaging about 40 km. per sec. for one component. there is also a general progression of velocity with brightness, the faintest stars having the highest average speed; and this is very much mixed up with the above dependence of speed on spectral type. to add to the confusion it has been found by plaskett that stars of tvpe o instead of having stil] smaller speeds than type b have moderately high speeds; and this is confirmed by the high speeds of planetary nebulae which usually contain a type o star as nucleus. eo it seems likely that the correlation of speed both with spectral type and luminosity results from a more fundamental dependence of speed on mass—the smaller the mass, the higher the velocity. it is tempting to suppose that there is equipartition of energy as among the molecules of a gas, the average value of 4 mv? being the same for all classes of stars. although the observations con- firm this to a certain extent, it is difficult to suppose that there is any theoretical basis for such a law. the mutual perturbations of the stars tend to bring about equipartition, but the process is too slow to be effective within the lifetime of the stars concerned. moreover, the empirical connection of mass and speed must not be pressed too far, because the o type stars which have very large masses have rather high speeds. the whole problem of stellar velocities is very mysterious. there must, we think, be a number of causes involved, and it seems clear that the stellar system has not yet reached anything like a steady and perma- nent distribution. | . moving clusiers—many years ago r. a. proctor pointed out a group of stars in the neighbourhood of the hyades with prac- tically equal proper motions; the researches of l. boss (*) have thrown new light on the nature of this association. he recog- nised as belonging to the group 39 stars spread over an area 15° square; the motions appear to converge towards a certain point in the sky—a perspective effect which would naturally occur if the actual motions in three dimensions are parallel; the direction of the convergent point gives the direction of the common mo- tion of the group relative to the sun. knowing the spectroscopic radial velocity of one or more members, we can by an easy geometrical construction find the whole linear velocity and also locate each star separately in space. we thus obtain exceptionally full and exact information as to the distances and luminosities of this group of stars. the cluster is roughly spherical with a diameter of 10 parsecs; there must be many non-associated stars—accidental interlopers in so large a region—and perhaps the most significant conclusion is that the casual attractions of these stars have not been able during the lifetime of the cluster to disturb appreciably the parallelism of the motions anc so scatter the cluster. another remarkable “ moving cluster ”’ is formed of five stars of the plough together with stars widely separated in the sky, including sirius, a coronae and 8 eridani. similar associations are specially frequent among the stars of the b type of spectrum, one of the most distinct being a chain of stars crossing the constellation perseus. astronomy 251 number and distribution of stars.—important statistics of the | the surface has to be maintained (otherwise the star would col- number of stars down to various limits of magnitude have been obtained by chapman and melotte and by p. j. van rhijn. we give some results of the latter investigation which is the more recent (24), the total number of stars down to photographic magnitude 16tm-o is 33,000,000; by a somewhat risky extrapolation it is estimated that the total number of stars in the system is between three and four thousand millions, and to reach half this number it would be necessary to go as far as magnitude 25\"-5. (exactly what is meant by the ‘‘ system ”’ in the foregoing sentence is somewhat difficult to define; there may, of course, be exterior galaxies or extensions which are not reckoned in these counts.) an important point is the well- known flattened distribution of the stars; up to magnitude 16tm the stars are distributed in the galactic plane 5} times as thickly as at the galactic poles. this is an increase compared with the con- centration of the brighter stars; up to magnitude 5tm, the correspond- ing ratio is 2}. we can easily understand this greater concentration of the faint stars, since on the average they carry us to greater distances, at which the oblate shape of the stellar system has more pronounced effect. taking a lower limit of luminosity 1/200 & sun, it is estimated that there are 30 stars within a sphere of five parsecs radius round the sun; about 20 of these have actually been identified. if this star density persisted, a sphere of 1,500 parsecs radius would contain 800,000,000 stars, besides an unknown but probably rather large number of extinct stars and of stars giving less than 1/200th of the sun’s light. this gives some idea of the possible extent of the star cloud to which we belong; there can be little doubt that the density must fall off very considerably within a distance of 1,500 parsecs, more especially in the directions of the galactic poles. the following table based on an investigation by kapteyn, van rhijn and weersma shows the average parallax of stars of different magnitudes:— ~—— mag. mean parallax mag. mean parallax 1tm-o 060\" 7m. 0090” m aa m (wilt 2tm.0 ‘o44 8-0 0065 bo “32° gtmo 0047” 4-0 0237\" 10tm-0 0034” ‘o titm-o 0025” ‘0 12tm-0 0018” it is an even chance that a particular star has a parallax between 0:83 and 1-13 times the average parallax for its magnitude. “volution of rotating masses.—the figures of equilibrium and the final disruption of rotating fluid masses have been studied in great detail by j. h. jeans. in agreement with liapounoff he has found that the so-called '' pear-shaped ”’ figure of equilibrium, which suc- ceeds the jacobi ellipsoidal form, is unstable. for a full account of his conclusions as to the evolution of double stars, spiral nebulae and clusters reference must be made to his book problents of cos- mogony and stellar dynamics (1919). with regard to the solar system, he finds himself unable to account for the formation of the planets by rotation alone; and he attributes them to a tidal disrup- tion of the sun having occurred at some distant epoch in the past. if this view is correct the system of the planets is a ‘‘ freak of nature,” owing its existence to a chance encounter of some larger star (which approached within less than the sun’s diameter from its surface). comparatively few other systems of this kind would be formed; and the common view that the stars in general are attended each by a system of planets may be entirely mistaken. age of the stars.—according to the helmholtz-kelvin theory, which supposes that a star’s heat is supplied by gravitational energy converted owing to contraction of the star, the age of the sun is not more than 30,000,000 years. physical, geological and biological arguments all agree that this time is far too short. an astronomical argument against the contraction theory is derived from the cepheid variables. like other giant stars 6 cephei would need to condense very rapidly if the heat which it squanders were derived from contraction; the increase of den- sity must in fact amount to 1% in 40 vears. since the period of the light variation is intrinsic, this must also change as the den- sity changes; calculation shows that the period should decrease 4o seconds annually. now 6 cephei has been under careful ob- servation since 1783 and the decrease of period is only just de- tectable—about 0-08 seconds per annum. thus in the present stage evolution is proceeding at no more than '/s0 of the rate required by the contraction hypothesis. in casting about for some other source of stellar energy, it is necessary to realise that no source is of any avail unless it liberates heat in the deep inte- rior of a star. the temperature gradient between the centre and lapse to the white dwarf state) and it 1s impossible to maintain a temperature gradient by supplying heat at the bottom end. this seems to rule out all suggestions that a star picks up its energy from outside. 7 granting then that the energy required for future radiation must be already within the star, we can at once set an upper limit to the amount. all kinds of energy possess mass, and the mass of the sun is equivalent to 1-8-10\" ergs of energy. this is the total store, and at the present rate of radiation it is sufficient for 15-10” years; actually it would last considerably longer since the rate of radiation would decrease as the sun’s mass decreased. to put the argument in another form, the mass of the sun’s radi- ation amounts to 130 million million tons per year; and the sun cannot go on losing this mass for more than the period stated because by that time there would be no mass left. with insig- nificant exception the store consists of energy of constitution of atomic nuclei and electrons, 7.e., subatomic energy; the whole amount can only be released if there is some process of annihila- tion of the electrons and protons which constitute matter. a less radical suggestion is that only transmutation of the elements occurs. in that case not more than !/1 of the mass can be re- leased as free energy; and to obtain even that much the mass must consist initially of hydrogen, since the main release of en- ergy occurs when hydrogen is transmuted into helium as a first step towards the synthesis of higher elements. the energy emit- ted in ordinary radio-active changes is insignificant from the cosmical standpoint. although the theory of annihilation of matter presents grave difficulties, it is attractive (1) because it is doubtful if the less radical hypothesis can give an adequate time-scale, (2) because stars observed in the earliest diffuse stage have on the average masses two or three times greater than the average mass of the more condensed stars, (3) because if a star does not change mass considerably all previous ideas of evolution of faint stars from bright stars collapse. a star might also change mass by picking up or losing matter from its surface, but calculations indicate that this effect must be small compared with the loss of mass by radiation. the sun’s chromosphere would have to move outwards at soo km. per sec. continually in order to carry off as much mass as is lost by radiation. | one difficulty is the co-existence of giant and dwarf stars in clusters. the larger the mass the faster is the rate of loss of mass; and however big the stars may be to start with, at the end of a million million years there can be no more than twice the sun’s mass left. in moving clusters such as the taurus cluster and pracsepe we find stars of more than twice the sun’s mass (judging from their luminosity) and therefore conclude that the age of these groups cannot be more than ro” years. but the loss of mass of a dwarf star in that time is almost negligible, so that the dwarf stars in these clusters must have been born with prac- tically their present masses. if we have to admit that the cluster dwarfs have undergone no evolution of mass, is there any point in assuming that dwarfs in gencral have been evolved from more massive stars? on the other hand, it mav be significant that these clusters appear not to have their full quota of dwarfs, there being few if anv below absolute magnitude about 7tm. very little progress has been made in unravelling the depend- ence of the rate of liberation of subatomic energy on tempcra- ture and density; but it seems to be established that the rate must increase with temperature or with density or both, since otherwise the stars would be unstable. it is also clear that there must be exhaustion effects, the older stars having used up their more prolific supplies. thus the sun liberates only 1-9 ergs per gr. per sec. as compared with 58 by capella, in spite of the fact that the sun is hotter and denser. a remarkable feature pointed out by ii. n. russell is that stars on the main series have nearly the same internal temperature. thus v puppis which requires a supply of yoo ergs per gr. per sec., and krueger 60 which requires 0-08, both have to rise to a central temperature of about .,0,000,- 000° to obtain their requirements. it looks as though 40,000,000° is a critical temperature at which subatomic energy becomes 252 available practically ad libitum; but this is scarcely credible as a physical hypothesis. according to kohlherster and millikan an extremely pene- trating radiation found in our upper atmosphere comes to us from outside the earth. if their experimental results are accepted there can be little doubt that this radiation originates in the nebulae or in the diffuse matter pervading space, and the great penetrating power points to subatomic processes as the cause of it. this indicates an evolution of the elements prior to the con- densation of matter into stars; and indeed astronomical evidence of the presence of complex elements in the most diffuse stars seems to require such an advance. but it baffles imagination to conceive how the four protons and two electrons required to form a helium atom can be brought together in a diffuse nebula. the whole problem of stellar evolution is now in a most chaotic state because it has become involved with questions of the laws of liberation of subatomic energy which as yet have proved to be too difficult for us. it should be understood that these at- tempts to form a theory of the subatomic generation of energy from astronomical studies, although unsuccessful, are not specu- lative in the ordinary sense of the term. when once it is ad- mitted that the stars’ radiation is provided by subatomic energy, it follows that the measurement of the liberation of this energy is one of the commonest astronomical observations; the attempt to co-ordinate the results of these measurements, and to dis- cover the laws connecting the rate of liberation with the physical conditions and age of the material is not airy speculation but a pressing problem of practical research. iti. clusters and nebulae globular clusters—about 70 globular clusters are known. (see 19.332.) they are distinguishable from the loose irregular clusters by their symmetrical and condensed appearance. a typical example, messier 3 (canes venatici), is shown in the plate, fig. 4, from a photograph taken at mt. wilson. the cepheid variables contained in some of these clusters have been used by h. shapley (*) to gauge their distances, relying on the fact that the absolute magnitude of a cepheid of given period is a known constant. in messier 3 the mean magnitude of 110 cepheids is 15tm-50, the individual stars devi- ating as a rule no more than otm-1 from this mean. in the cluster w centauri 76 cepheids concentrate with similar closeness about a mean magnitude 13tm-57. it is clear that the difference 1tm-93 must correspond to the greater distance of messier 3; and it is easily deduced that the ratio of the distances is 2:43, this ratio being very accurately determined provided that absorption of light in interstellar space is negligible. we are not quite so certain of the absolute distances of the two clusters; but the evidence seems to indicate that the absolute magnitude of cepheids with periods less than a day is—otm-2, which gives the following distances—w centauri, 5,800 parsecs; messier 3, 14,000 parsecs. when it is recalled that the usual trigonometrical method can scarcely be applied to determining distances greater than 50 parsecs, the extraordinary power of this method of plumbing space will be realised. the method was first used by e. hertzsprung to determine the distance of the lesser magellanic cloud. by this method, and by supplementary devices, shapley has been able to plot the distribution of the globular clusters in space and to form an idea of the extent of the system which they out- line. even in this vaster system the galactic plane is still a plane of symmetry and of flattening, though the clusters extend to great distances above and below, the average distance from the plane being eight kiloparsecs. in plan the system is elongated with its axis in galactic longitude 325°—nearly the direction of star streaming; the greatest diameter is at least 60 kiloparsecs, and the sun is near one end of it, so that practically all the globu- lar clusters are found in one hemisphere of the sky. the most remote cluster known is distant 67 kiloparsecs or 200,000 light years. we have to recognise that the “ stellar system,” dealt with in the researches described previously, is but a small star cloud in this greater galactic system. roughly speaking those astronomy researches may be considered to relate to a domain of about 800 parsecs radius; the sun seems to be fairly centrally placed in the local star cloud {about 90 parsecs from the centre according to charlicr), but this is on the outskirts of a greater system whose centre is 20,000 parsecs away. | spiral nebulae.—the idea that there may exist “island uni- verses ” independent of and co-equal with our own galactic sys- tem of stars has appealed to many astronomers from time to time. if any known celestial objects deserve this name it must be the spiral nebulae, and after many vicissitudes this theory of their nature is now most favoured. it is natural to suppose that our flattened system of stars with the clusters and obscuring clouds of the milky way coiled round, would if seen from a dis- tant point, look rather like a spiral nebula such as that shown in the plate, fig. 3 (m. 81, ursa major). the spiral nebulae show no tendency to crowd towards the galactic plane. most of them have very large velocities of recession of the order 1,000 km. per sec., which seems to argue a lack of dynamical association with the galactic system. the great difficulty has been to know how far away we must place them. rude estimates of distance based on the brightness of the novae which have sometimes been discovered in them suggested that they were extremely remote. on the other hand a. van maanen believed he had measured comparatively large proper motions of points of the nebulae, which showed that they were not so remote as the glob- ular clusters; but it was impossible to reconcile these results with any consistent scheme of mass and dimensions of the nebu- lae, and moreover the rotation shown by the supposed proper motions was in the opposite direction to that shown by the radial velocities. in 1924 e. p. hubble discovered and observed a number of cepheid variables in the great andromeda nebula, and using these as a gauge, he deduced a distance of 300,000 parsecs (a million light years). presumably the smaller spirals are still more remote. this seems to settle the question of dis- tance and size of these nebulae. accepting hubble’s result, the andromeda nebula is still inferior in size to the galactic system of stars and globular clusters, but it is at any rate an object of comparable class, and is fairly entitled to be regarded as an island universe. the theory of relativity suggests an explanation of the observed high receding velocities of the spirals. according to de sitter’s theory of a curved space-time remote objects would be expected to show a displacement of their spectral lines to the red, partly owing to a “slowing down of natural processes ” in remote regions and partly owing to genuine receding velocities acquired under the cosmical repulsion which exists in a world of de sitter’s type. this explanation is, however, very much weakened by the fact that we know of two spirals (the andromeda nebula being one of them) approaching with speeds of 300 and 260 km. per second. gaseous nebulae.—other classes of nebulae are comparatively near objects within the ordinary limits of the stellar system. the diffuse nebulae are of gaseous constitution or possibly a mixture of gaseous and meteoric matter. since they occupy great tracts of space their densities must be extremely low, other- wise they would have large masses which would have conspicu- ous effect upon the distribution of stellar velocities. besides the bright diffuse nebulae we recognise dark nebulae or obscuring patches, which cut off the light of the stars behind and cause apparently void regions in the sky. it is not always possible to tell whether a void region is due to one of these patches or is a genuine lacuna in stellar distribution, but there are many cases where the existence of dark nebulosity is undoubted. it is be- lieved that bright and dark nebulae are of the same nature, the luminescence of the former being due to stimulation by the radia- tion from the stars present in them. thus the orion nebula is caused to shine by the numerous hot b type stars in that con- stellation; a similar nebulosity in taurus is dark because there are no b type stars present. the connection of nebular light with stellar stimulation is well seen in hubble’s variable nebula; here there is one principal star which is a variable, and the nebula also varies. owing to the low density and the effect of ultra- athens violet radiation from the stars the material of the bright nebulae is rather highly ionised; this probably accounts for the fact that most of the lines of the nebular spectrum are unidentified, the conditions being difficult to reproduce in the laboratory. of the identified lines, hydrogen, helium, ionised helium and prob- ably carbon are found. planetary nebulae are disk- or ring-shaped clouds which sur- round a single central star. the star is always found to be extremely rich in ultra-violet ight, and is probably to be classed as type o. it is sometimes thought that this aureole consists of atoms expelled and sustained by radiation pressure; but on closer examination we find it difficult to see how radiation pres- sure can have much to do with the phenomenon. these nebulae are found by spectroscopic observation to be rotating, and there can be little doubt that rotation plays a fundamental part in the equilibrium. fig. 1 on plate, due to w. h. wright, shows a photo- graph of the ring nebula in lyra taken through a prism, so that monochromatic images are shown for each of the chief emission lines. it will be seen that the different wave-lengths give rings of different sizes, and the great difference of distribution is well shown by the bright images on the extreme left and right (both corresponding to unidentified spectral lines). this may be due either to actual differences in distribution of the gases giving these lines, or more probably to differences in the conditions of ionisation. calcium cloud in space—in certain spectroscopic double stars the phenomenon of “ fixed calcium lines ” is observed. whereas the other spectral lines shift to and fro as the star approaches and recedes in its orbit, the h and k lines of ionised calcium remain stationary. it is clear that somewhere between us and the star’s photosphere there must be an absorbing cloud of calcium vapour which does not follow the star in its orbit. the phenomenon was first pointed out by hartmann in 1904 for the star 6 orionis. later miss heger discovered that the d lines of sodium also remain fixed in 6 orionis. the same be- haviour has now been observed in a large number of stars, but no other “ fixed ” spectral lines have been found. the impor- tant question to decide is whether the cloud belongs to the double star or whether it is a continuous cloud filling interstellar space. j.s. plaskett (26) has shown that the latter alternative 1s correct; the motion of the calcium cloud is often different from that of the centre of mass of the star. after removing the solar motion the velocity of the cloud relative to the mean of the stars is found to be small. just as there are lines in the solar spectrum which do not share in the sun’s rotation and are accordingly to be attributed to absorption during the passage of the light through the earth’s atmosphere, so we have fixed lines of calcium and sodium which do not share in the orbital or individual motion oi the star and are to be attributed to absorption in an interstellar ‘‘atmosphere.”? the fixed lines only appear in the spectra of the hottest stars, but that is perhaps due to the fact that cooler stars have strong h and k lines of their own, masking the lines of the cloud. it has been suggested that the presence of the hot star is necessary in order to ionise the calcium vapour and render it capable of absorbing h and k light, so that although the ab- sorption is performed by the interstellar cloud only the parts of ‘the cloud stimulated by the star are effective. but this explana- tion would not hold good for the d lines of sodium which are absorbed by un-ionised unexcited atoms. there can be little doubt that the absorption occurs equally along the whole track of the light through space, and the intensity of the lines should be an indication of the length of track, that is the distance of the star. in order to avoid a huge mass of the stellar system inconsistent with the observed velocities of the stars, it must be postulated that this interstellar cloud is of very low density. about one atom per cubic centimetre is the maximum that can be allowed. it is calculated that matter so diffuse as this would take up a high temperature not much lower than the photospheric tem- peratures of the stars; although a black body in interstellar space would sink to a temperature of 3° absolute. the density is too small to give any appreciable scattering or absorption of light 253 in space other than the special line-absorption of calcium and sodium, references.—(1) g. e. hale, astrophys. jour., vol. 28, p. 315 (1908); (2) zbid., vol. 62, p. 270 (1925); (3) tbid., vol. 38, p. 27 (1913): f. h. seares, observatory, vol. 43, p. 310 (1920); (4) c. g. abbot, smithsonian misc. coll., vol. 77, no. 5 (1925); (5) e. a. milne, monthly notices, vol. 85, p. 111 (1925), vol. &6, p. 8 (1925); (6) pub. ast. soc, pacific, oct. 1924; (7) e. w. brown, yale obs. trans., vol. 3, pts. 1 and 3; (8) h. glauert, monthly notices, vol. 75, p. 489 (1915); (9) j. k. fotheringham, monthly notices, vol. 80, p. 578 (1920); (10) lick obs. bull., no. 355; (11) trans. inter. astr. union, vol. 1, p. 71 (1922); (12) j. jackson and h. h. furner, afonthly notices, vol. 81, p. 2 (1921); (13) f. g. michelson and a. a. pease, astrophys. jour., vol. 53, p. 249 (1921); (14) a. s. eddington, monthly notices, vol. 84, p. 308 (1923-4); (15) w. s. adams, proc. nat. acad. sc1. (july 1925); (16) m. n. saha, proc. roy. soc., vol. 99, sci. a, p. 135 (1921); (17) w. s. adams, etc., astrophys. jour., vol. 53, p. 13 (1921); (18) h. shapley, astrophys. jour., vol. 40, p- 448 (1914); (19) a. s. eddington, monthly nottces, vol. 79, pp. 2, 177. (1919); (20) a. s. eddington, 7ibid., vol. 77, p. 4 (1gii); a. s. eddington and w. e. l. hartley, zi7d., vol. 75, p. 521 (1915); (21) k. schwarzschild, gettinger nachrichten (1907-8); (22) oort, bull. ast. inst. netherlands, no. 23; (23) l. boss, astren. jour., no. 604 (sept. 25 1908); (24) van rhijn, groningen pub., no. 27; (25) h. shapley, astrophys. jour., vol. 48 (1918); (26) j. s. plaskett, monthly notices, vol. 84, p. 80 (1923-4). for astrophysical problems two up-to-date books are available— f. j. m. stratton, astronomical phystcs (1925) and h. dingle, modern astrophysics (1926). other books relating to various branches of the subject that may be found useful are: a. s. eddington, stellar movements and the structure of the untverse (1914); r. g. aitken, the binary stars (1918); j. h. jeans, problems of cosmogony and stellar dynamics (1919); 5s. newcomb and r. engelmann, populdre astronomie (6th german ed., 1921); h. s. jones, general astronomy (1922); c. h. payne, stellar atmospheres (1925); a. s. icddington, ternal constitution of the stars (1926). for technical information as to the nebulae, lick obs. pudb., vol. 13 (1918) should be consulted. (a. s. e.) athens (sce 2.831).—in spite of continual war between 1912 and 1922, athens has progressed in every way since 1910. as the seat of government, it is also the focus of society, politics and trade. its port, piraeus, the best equipped and principal harbour of greece, is the terminus of the railway and steamship lines, and the mart for all imports and exports. as new quarters have been laid out to accommodate the rapid growth of population, and the refugees from asia minor, athens now extends all round the base of mount lycabettus. outlying parts, such as patesia and pangrati, have enormously increased, and suburbs such as ke- phisia, marousi, phaleron and kallithea have much expanded. new discoveries —its museums make athens an art centre of the first rank, and its monuments always provide fresh material for our knowledge of hellenic culture. the constant study, which scholars of all nations devote to the acropolis, takes practical shape in such work as the re-erection of fallen portions of the erechtheum, propylaea and parthenon. recent discoveries show that the odeum of pericles was a rectangular hypostyle build- ing and not circular, as formerly supposed, and that the seats of the theatre of dionysus in its earliest form were in straight lines. an accident revealed two archaic sculptured bases built into a later wall, perhaps that of themistocles, not far from the dipylon. these, which still retain some of their original colour- ing, are sculptured with chariot groups and athletic scenes. in their freshness and simplicity they are charming examples of archaic attic art. so the excavation of the agora, now planned by the americans, may well lead to most important results. the area to be cleared, which is bounded on the west by the temple of hephaestus, on the east by the horologium of andronicus, and on the north by the stoa of attalus and the library of hadrian, is now covered by narrow streets and small houses, many of which date from turkish times. when greece became free, ross, the first head of the department of antiquities, proposed that this should be re- served, and, when excavated, become a public park. round the agora were colonnades, the stoa basileios and the painted stoa which had frescoes of the battle of marathon, the temple of the mother of the gods and other shrines. public buildings here were the bouleuterion, or senate house, the prytaneion, the centre of state hospitality, and the tholos, 254 or round building. monuments and statues of all kinds, com- memorating distinguished citizens or foreigners and great events, were erected among them, and also inscriptions with official records and accounts. the discovery of one of these buildings, with its art treasures and original documents, for greek history would be an event of first-rate importance. (a. j. b. w.) athletics (sec 2.846).—the london olympiad of 1908 may well be regarded as marking the commencement of a fresh athletic era throughout the world. it is worthy of note that the united states, in pursuit of a progressive policy, has always at once included in the amateur athletic union and inter- collegiate championship programmes any new event which might be added to the olympic syllabus. the english amateur athletic assn. c(a.a.a.), on the other hand, for many years ignored such events as throwing the javelin, discus and 56 ib. weight and the hop, step and jump, and allowed to fall into clisuse, through lack of encouragement and competition, such exccllent exercises as pole vaulting, shot putting and hammer throwing, and gave but little more attention to high and long jumping and hurdling. in roro the english amateur field events assn. (a.f.e.a.) was formed and authorised by the a.a.a. to hold champion- ships. by 1914 the purpose of the a.f.e.a. had been ful- filled, for in that vear the a.a.a. incorporated in its champion- ship programme the javelin, discus, hlop, step and jump and the 440 yards low hurdles, but even then these events were not taken as seriously as the others. fifth olympiad.— meanwhile the holding of the fifth olympiad (1912) had been allotted to stockholm and the swedes had recalled from america that great athletic coach, ernie hjertberg, to make ready a national team, finland, also, had produced a great distance runner in iannes kolehmainen, and a set of magnificent heavy weight field events men, such as saaristo, taipale and niklander, while france had come into prominence with jean bouin, a world’s record holder, who was, however, beaten by kolehmxinen in both the 5,000 and 10,000 metres races. germany, too, gained prominence with rau, the sprinter, and braun, the middle distance runner. it is significant that at the stockholm olympiad the united states finished first with a total of 85 points, finland second with 29 points, sweden third with 27 points and great britain fourth with 153 points. american athletes were again in the ascendant, but with their supremacy challenged by finland and sweden. finland, sweden, norway and denmark had now be- come definitely athletic countries, while france, germany and the lesser european nations, such as italy, belgium and hol- land, were all showing steady progress toward national athletic efliciency. 3 afore international afaiches—about this time, too, the custom of holding international athletic matches became pop- ular. most notable of all these, perhaps, is the scandinavian landskamp, in which norway, sweden and denmark mect annually at oslo, copenhagen and stockholm in rotation. france, too, entered the international arena with matches with sweden and belgium, and great britain began to hold a triangular international in which england, ireland and scotland met annually. ‘the other european nations in meeting each other contest practically the whole of the clympic programme, whereas from england's match with france are excluded such important events as javelin throwing, pole vaulting and the low hurdies, and in the case of the british triangular inter- national the discus is also omitted. in any comparison of international prowess in field events it is only fair to note that neither discus nor javelin throwing is practised at oxford and cambridge, and the hammer throwing event has been aban- doned, while these events are regular features at all american and foreign schools, colleges and universities. the war poi atte stockholm came the world war, which prevented the sixth olympiad, although a vast stadium had been built to house it at the grunewald, berlin. many doubtless thought that in those strenuous years all sport must athletics come to an end. this was not to be. alva kranzlein, the ameri- can sprinter-hurdler-jumper, had returned to germany, the land of his forefathers, to make ready the german olympic team, and in germany or holland he stayed throughout the war, laying the foundations of a great athletic future for those nations. ile was probably the first person to discover, in the internment camp at gravenhaag, holland, the potentialities of h. f. v. edward, the west indian, who won so many a.a.a. championship files in england championships were abandoned from 1914 to 19t9. athletic meetings of a sort continued to be held under an unofficial general amnesty, which allowed pure amatcurs and those soldiers who had forfeited their amateur status to compete together. in great britain the london athletic club contrived to carry on the public schools sports meeting right through the war, thus assuring for great britain the nucleus of a fine supply of athletes of international standing when the days of war should be ended. education of athletes —here, one may pause to draw atten- tion to the splendid movement now in progress all over the world for the better athletic education of boys at school. in the united states the universities and colleges, as well as every school of any standing, have their properly qualified athletic coach. inter-university, inter-collegiate and inter-school athletic meets are exceptionally popular, and the same may be said of the scandinavian and many other european nations. france, in particular, has established a ministry of sport. in scotland inter- scholastic championships have long been in vogue. in england the public schools sports meeting and also the public schools relay meeting, the latter under the auspices of the achilles club, are both established fixtures, and there is now an inter- schools athletic assn., which, for the first time, held champion- ships in 1925. the public school athletic league of new york, organised in 1906, annually has between 7 and 8 thousand boys compete in its championships, all on one day. it is a usual thing to have between 2,500 and 3,000 boys compete in the indoor elementary championships each year. but in no country, save possibly the united states, is there to be found anything ap- proaching the swedish schoolboys athletic week, held annually in stockholm, to which city school teams journey from every part of sweden. the war period again.—to revert, however, to the war period. the united states, unaffected by the struggle in its early stages, and sweden, norway and denmark, which coun- tries maintained their neutrality throughout this time of strife, forged rapidly ahead. in rors at cambridge, mass., norman 58. vaber placed upon the books a new amateur mile world’s record of 4 min. 12° sec., which, at last, eclipsed the profes- sional mile time of w. g. george, who, in 1866, covered the distance, in his match with w. cummings at i. illy bridge, in 4 min. 12} seconds. taber's race was the forerunner of a great many record-breaking performances. in 1916 at evanston, u.s.a., r. [. simpson brought the 120 yards migh hurdles, record dow n to 142s sec., owing to certain modifications he made in the, then, accredited style of hurdling, and in 1920 a young canadian, earl thomson, who had served in the roval air force, still further reduced the record time to 14°% seconds. two of the most noteworthy achievements of the war period were the performances accomplished in 1916 by j. e. meredith (u.s.a.), who sect up new world’s records fer the quar ter mile, which he ran in 472% sec., and the half mile run in rmin. 5 asin seconds. at magdeburg in 1913, incidentally, a. r. taipzle (f inland) threw the discus 158 ft. r1 inches. ‘this per- formance is duly recognised in scandinavia, but has never been passed by the international amateur athletic federation; other- wise it would stand as the world’s record. shortly after the signing of the armistice a great inter- allied military athletic meeting took place in the pershing stadium at paris. of signal importance at this time was the step taken by the authorities of the british services, who decided that the pre- war custom of rewarding athletic proficiency among soldiers and sailors by money prizes must forthwith cease. an inter- athletics services meeting was held in rgr9, at which some of great britain’s dominion soldiers, not yet demobilised, proved clearlv that the dominions themselves would hold a strong hand at the next celebration of the olympic games. it is interesting to note that even after the united states had entered the world war, the national a.a.u. championships were not abandone:, whereas, in great britain, no championship meetings teok place between ror4 and rgro. peace conditions return.—in the year of the restoration of the a.a.a. championships the governing body again elected to omit certain of the ficld events from the programme; nor have they since insisted upon the inclusion of these events, which score, equally with the track events at the olympic games, in such im- portant contests as the triangular international between england, scotland and ireland, the annual match between england and france, and the inter-county championships. the governing body ini925 decided to abandon the english national cham- pionships, after they had been held for only three years, and were just beginning to produce a really satisfactory crop of fine young english exponents of these field events. the a.a.a. championships are open to the whole world and hitherto great britain has always been outclassed at the a.a.a. championships proper in the majority of field events in great britain the sterling post-war work of the oxonians, a.n.s. jackson and b.g.d. rudd, coupled with the exertions of the cambridge men, p. j. baker, g. m. butler, r. s. woods and w. r. seagrove, was responsible for inducing university athletes to take a healthy interest in open competitions generally, and championship meetings in particular, outside the limited scope of their own university sports. in 19019 the inter-university athletic board of great britain and irelund was constituted, comprising the universities and university colleges of aberystwith, bangor, birmingham, bristol, cardiff, durham, leeds, liverpool, manchester, not- tingham and sheftield; other universities have since joined the movement. at these provincial universities the whole of the olympic events are practiscd, and year by year the records, especially in the ficld events, improve. seventh olympiad—it was decided to hold the seventh olympiad at antwerp in 1920, partly as a tribute to the belgian people for the part they had played in the war, but principally in order that the true glympic cycle might not be interrupted. there was a great outcry that the war-worn nations were not yet sufficiently recovered to participate in such a festival. p. j. baker, cambridge university, was appointed captain of great britain’s athletic team; the whole olympic side came under the control of brig.-gen. r. j. kentish, c.m.g., d.s.0o., and the british team won golden opinions in antwerp. great britain did better in actual competition at this olympiad than she had ever done at any previous celebration of the games. notable victories were gained by a. g. hill in the 800 and 1,500 metres, by b. g. d. rudd, the oxonian, representing south africa, who won the 400 metres, and by percy hodge, who broke the world’s record in the 3,000 metres steeplechase. the british team succecded also on this occasion in winning the 1,000 metres relay race. at the conclusion of the antwerp games the united states was first, finland second, sweden third and the british isles fourth, the same order as appertained at the conclusion of the stockholm games, but in the other positions there was a marked difference. france, for example, eighth at stockholm, was now fifth, while italy had moved up from the eleventh to the sixth place. america’s strength lay in the sprints, hurdles, relay races and jumps. finland gained honours across country, in the middle distance races and the throwing events. sweden scored heavily through the magnificent team work of her men, and italy came into prominence by the fine walking of ugo frigcrio. further proof of the taste which the public was acquiring for international competition was exemplihed immediately after the games by the match between france, sweden and the united states, which took place in paris, and that even greater match at quecn’s club, london, between the united states and the 255 british empire. this match has evidently come to stay, since it was repeated in london in 1924, aiter the celebration of the eighth olympiad at paris. in ro21 the international amateur athletic federation held an important congress at geneva. new rules for international competitions were passed and the olympic programme at last standardised. the years that followed the antwerp olympiad were years of wonderful progress. this is conclusively proved by the existing world’s records | running.—in 19006 the 100 yards record of 935 sec. was cstab- lished. this record was equalled in 1922 by the canadian, c. coaffee, while c. w. paddock, u.s.a., has also, upon several occasions, covered the distance in that time. paddock has also, in r921r and 1924, run 220 yarils in 204% sec., thereby establishing himself as the greatest sprinter the worl has ever seen. ‘taber’s mile record, to which reference has already been made, was, in 1923, quite eclipsed by the finnish phenomenon, p. nurmi, who, in a race at stockholm, beat the great swede, edvin wide, in a mile race run in 4 min. 102% seconds. from one mile up to 10 miles the records are about evenly divided between nurmi and alfred shrubb. earl thomson still holds the palm among high hurdlers, but the 220 yards low ilurdles record of 231% sec. was made by c. e. brookins (u.s.a.) in 1923, and that of 5345 sec. for the 440 yards low ilurdles by i. h. taylor in 1925. jumping—in jumping the most marvellous improvements have been made. old-time sportsmen still remember the amaze- ment that was felt when, in 1876, m. j. brooks of .oxford university accomplished a high jump of 6 ft. 25 in., and again in 1893 when that other great oxonian, c. b. fry, created a world’s long jump record of 23 {t. 6} inches. the latter performance was totally eclipsed by the irishman, p. j. o’connor, who, in 1901, cleared 24 it. 11} inches. credit for the subsequent improvement in jumping belongs to america. taking the high jump first, it must be remembered that in 1895 m. j. sweeney (u.s.a.) cleared 6 ft. 5§ in., and that by ro2z an englishman, b. howard baker, had cleared 6ft. 5 inches. in the meantime, however, george horine, a young undergraduate of leland stanford university, california, had evolved an entircly new method of jumping, styled the “western roll,”? by which, in 1912, he cleared the height of 6 ft. 7 inches. in 1914, however, edward beeson, another american, reached 6 it. +5. in., and even this has been exceeded by yet another amer- ican, h. m. osborne, who, in 1925, jumped 6 ft. 84 in.; this was no fluke performance, for he has, time and again, gone within an ace of beating even this stupendous record. from the beginning of the present century, when o’connor set the world’s long jump record at a quarter of an inch short of 25 ft., it was a fruitful source of discussion as to whether such a jump could be made again and even 25 ft. possibly be beaten. at stockholm in r9r2 albert gutterson (u.s.a.) only just failed to reach o’connor’s mark, his actual jump measuring 24 ft. 11% inches. early in 1920 the united states began to produce a crop of most amazing negro jumpers. sol butler is said to have beaten os ft., but the record never went on the books. in 1921, how- ever, fe. o. gourdin, harvard university, actually jumped 25 ft. 3 inches. shortly after this, yet another negro, de hart hub- bard, repeatedly beat 25 ft, and in exhibitions cleared 25 ft. 4 in.; being exhibitions, however, these jumps did not stand as records. at the paris olympiad de hart iiubbard pulled a muscle and, although winning the competition at 24 ft. 5y% in., was far below his best form, and it was left to r. legendre, a white “american, who was competing in the pentathlon, to carry the world’s record 6n to 25 ft. 58 inches. since then de hart hub- bard has attained his ambition by clearing 25 ft. 10} inches. for many years it was believed impossible that a pole vaulter would ever beat 13 feet. at antwerp in 1920, however, i’, foss (u.s.a.) accomplished 13 ft. 5 in. in a downpour of rain, and in 1922 charles hoff, of norway, at copenhagen cleared 13 ft. 6 inches. again at copenhagen, a year later, the same athlete cleared 13 ft. 92 inches. it was confidently anticipated that hott would beat 14 ft.at the paris games, but just prior to the olym- piad of 1924 he broke his ankle and so could not compete. in 1924 it was reported that r. e. spearrow (u.s.a.) had, at tokio, 256 cleared the amazing height of 13 ft. ro} inches. this record has not yet been accepted, but at abo in 1925 hoff, the norwegian champion, actually cleared 13 ft. 1142 inches. it is significant that nations that have hitherto been con- sidered non-athletically disposed, such as the south american peoples and the japanese, are really taking seriously to sport. at the olympic games at paris in 1924, for instance, when a. w. winter, new zealand, set up a new world’s hop, step and jump record of so ft. 1133 in., l. bruneto, of argentina, was second at so ft. 74 in., which easily beat the previous olympic record, while m. oda, japan, was sixth at 46 ft. 9 inches. weight putting —the world’s record for putting the 16 lb. shot has stood at 51 ft. to the credit of ralph rose (u.s.a.}, since 1909, but has since been closely approached, both by other americans and by finnish and swedish men. the record of 189 ft. 64 in. for throwing the 16 lb. hammer was created in 1913 by p. ryan (u.s.a.) and has never since been seriously challenged. it is interesting to note, however, that the javelin throwing record which, in 1908, stood to the credit of e. v. lemming, sweden, at 179 ft. 103 in., was raised to 216 ft. 102 in. by j. myrra, finland, in tgo19, and again eclipsed, with a throw of 218 ft. 74 in., by h. lindstrom, sweden, in 1924; this thrower has since done 221 ft., which record has not, however, been accepted by the international amateur athletic federation. still more interesting is it to note that, in 1924, the british thrower, j. dalrymple, accomplished 186 ft. 5 in., while a number of am- ericans have exceeded 200 ft. and the frenchman, degland, has beaten 190 feet. from 1912 until 1925, the world’s official discus throwing record of 156 ft. 12 in. was held by j. duncan (u.s.a.). in sept. 1924 thomas j. lieb (u.s.a.) broke this record with a throw of 156 ft. 23 inches. clarence “ bud” houser (u.s.a.), the olympic shot and discus champion, has bettered 158 ft. twice in 1925. this performance was closely approached upon several occasions by the two finnish throwers, taipale and niittymaa, and has been eclipsed by glenn hartranft (u.s.a.), who, on may 2 10925, reached 157 ft. 18 inches. marathon running —marathon running, at the official dis- tance of 26 m. 385 yd., is an event competed for but once or twice annually in any country, by reason of the strain imposed upon the runners. but here again a great improvement has taken place during the decade 1915-25. in rg08 dorando pietri, the little italian waiter, ran from windsor to london in 2 hr. 54 min. 463g seconds. at stockholm in 1912 the course was only about 25 m. and the time taken by the winner, k. k. mcarthur, south africa, was 2 hr. 36 min. 5446 seconds. at antwerp, however, the full course was covered and the race won by h. kolehmainen, finland, in 2 hr., 32 min. 354 seconds. in 1924 numerous olympic trials were held in the united states and one of these was won by clarence de mar, in what must be considered world’s record time of 2 hr. 29 min. 40} seconds. the actual paris marathon was won by a. qo. stenroos of finland in 2 hr. 4r min. 223 sec., r. bertini, italy, being second, and clarence de mar, united states, third. cross couniry—it has been said that the home of cross country running is england and to this form of recreation the country owes the long-continued successes of so many of her men in the middle distances upon the track. englishmen have, how- ever, to-day found serious rivals in this branch of sport in the french teams that annually visit the country, in the finns who have always beaten them at the olympic games, and also in the united states, where the cult of cross-country running has ob- tained a firm hold. switzerland, too, is becoming a recognised force in international competition. at the paris olympiad in 1924 j. imbach for a brief space held a new world’s 400 metres record, while p. martin, switzerland, finished second to d. g. a. lowe, great britain, in the 800 metres, and w. scharer, switzerland, was second to p. nurmi, finland, in the 1,500 metres. walking —walking is a sport which has had a certain vogue in great britain, america and canada, but which is more en- thusiastically practised in italy than elsewhere. the great days of walking in england were those in which g. e. larner, a athletics brighton policeman, and g. j. webb, who had been both sailor and soldier, were completely unbeatable at the 1908 olym- piad, in which they took first places in both the 3,500 metres and the 10 miles walking events.. before the stockholm olympiad, larner had retired and it fell to webb to uphold england’s honour at 10,coo metres, the only walk included in the pro- gramme. webb, however, was beaten by g. h. goulding of canada, but finished ahead of e. l. altimani, of italy. at antwerp in 1920 the britishers were completcly outclassed by ugo frigerio, who won both the 3,000 and 10,000 metres events, and also by the two men from the dominions, g. l. parker (australia) and c. c. mcmaster (south africa), as well as by the two americans, r. f. remer and j. b. pearman. at paris, in 1924, england again suffered defeat, frigerio finishing first in the 10,000 metres walk, followed by g. r. goodwin (great britain), c. c. mcmaster (south africa) and his fellow italian, d. pavesi. the world’s walking records are as follows: g. e. larner (great britain), 2 to 6 miles and 8 and 9 miles, and g. h. goulding (canada), half mile, 1 mile and 7 miles. all other records from 11 to 25 m. are held by british walkers. at the eighth olympiad held in paris in 1924, at which nearly 2,000 athletes representing 45 different nations took part in the track and field events, the united states, in 27 events, scored 12 first places, made 5 of the 9 new world’s records and estab- lished 2 of the 5 new olympic records, while one of the two marks, equalling previous olympic records, was also established by an american. of the other countries finland scored 9 first places, great britain 3, new zealand 1 and italy tr. athletics in great briiain.—in america, scandinavia, on the european continent and even among the coloured races of the world, athletic progress of an amazing kind is taking place from year to year, but up to 1925 it seemed certain that great britain must fall so far behind as to be at last forced to abandon the struggle altogether. the difficulty, however, was overcome when it was realised that hitherto british athletic sport had been too individualistic to appeal to the british sporting spirit. the decentralisation of authority was decided upon and county amateur athletic associations began to come into being through- out the country. the essence of the county administrative scheme js found jn the internal management of the sport by counties within their own areas, inter-county contests being an essential adjunct to the movement. at present bedfordshire, which county held matches in 1925 with the london athletic club and the univer- sity of london a.c., has probably achieved the most signal progress. in this connection it may be stated that england has for years past been divided into northern, midland and southern areas for administrative purposes, and that the furtherance of the county scheme lies at present entirely in the hands of the south, a circumstance strongly resented by the northern and midland districts. in 1925 the first english inter-county relay and team athletic championship, for the trophy presented by the achilles club, was held at stamford bridge, london; and middlesex proved the winner. the position, however, was an entirely un- satisfactory one. the fatal policy of booming certain events at the expense of others was once again fully in evidence, such events as hammer, discus and javelin throwing, the pole vault, hop, step and jump, and the 440 yards low hurdles being excluded from the programme. the result was that bedfordshire, exceptionally strong in the field events and hurdles and several other counties refused to take part in championships which were not considered fully representative of english ath- letic sport. a pronouncement was made subsequently that the championship programme would not be in any way aug- mented in 1926. consequently the midland counties, comprising bedfordshire, gloucestershire, leicestershire, shropshire and staffordshire, for the time being, abandoned all thought of taking part in the inter-county championships. the north also, although possibly from different motives, refused to have anything to do with the county scheme. in deference to the views of the coun- ties, however, it was finally agreed to include pole vault and atlanta—atmospheric electricity throwing the discus, but this still leaves hammer and javelin throwing outside the scheme. the essential fact is that, up to 1906, the amateur athletic assn., which is the governing body of the sport in england, has been unable to establish any liaison with the english public schools, which should prove the great recruiting ground for future olympic teams. on the other hand, the counties, by reason of their more personal local contact, have in one year begun to till this field, as is witnessed by the number of schoo!}- boys who gained their county athletic colours during 1925, and the far greater number of school authorities who sought the ad- vice and assistance for coaching schools for the annual sports, of old champions and other experts. women as athletes ——one aspect of international athletics still remains. it is the surprising degree of efficiency which women are attaining in this particular phase of sport. long before the war women’s colleges in the united states recognised and practised track and field athletics keenly and also kept records; but it was only during, and to a greater extent since, the war that british, french, czechoslovakian and belgian girl athletes took up ath- letics seriously. mrs. sophie eliott-lynn, whose sporting record ranges from big game hunting to aviation, and mrs. vy. cambridge, have been the prime movers in great britain and their efforts have been ably seconded by messrs. lyons’ ladies’ a.c., the london olym- piades and the middlesex ladies’ athletic club. it may be said, indeed, that the women’s athletic movement started in england and has spread outwards. as was natural, the records remained chiefly in the hands of english girls during the first years follow- ing upon the inception of the scheme. championships, inter- national matches and a great international festival, akin to the men’s olympic games, are now held annually; performances are improving and records being broken week by week. in 1924 a team of czechoslovakian girls met and was defeated by an english team. in 1925 the interest of this international match was greatly enhanced by the arrival in london of a team of canadian girls. | before the triangular international meeting in 1925 between great britain, canada and czechoslovakia the world’s records stood as follows: 100 yards, rose thompson (england) 1126 sec.; 250 metres, e. edwards (england) 34 sec.; 100 yards hurdles, g. sabie and f. batson (america) 14% sec.; 440 yards, v. palmer (england) 61 3¢ sec.; 880 yards, m. lines and i. trickey (england) 2 min. 263% seconds. records in the throwing events represent the best right and left hand throws added together and are as follows: putting the 8 ib. shot, v. morris (france) 68 ft. 11 in.; throwing the javelin, mlle. vidlakova (czechoslovakia) 161 ft. 24 in.; throwing the 3} ib. discus (best hand only), v. morris (france) 98 ft.; high jump, phyllis green (england) 5 ft.; long jump, mile. mejzlikova (czechoslovakia) 17 ft. 424 inches. at the women’s international held at stamford bridge, london, on aug. i 1925, miss e. trickey (england) set up a new half mile wom- en’s world record of 2 min. 24 sec., while miss v. palmer, also of the english team, beat the world's record in the 250 metres, which distance she ran in 334% seconds. this movement, which was almost purely british in its original conception, is now gaining a true international vogue, and, towards the end of 1925, a team of british girl athletes journeyed to scandi- navia for the purpose of assisting at the inauguration of a swedish women’s amateur athletic assn. and also took part in an inter- national meeting at stockholm. (see olympic games.) bibliography.—j. a. mussabini, the complete athlettc trainer (1913); f. a. m. webster, olympian field events (1913); evolution of the olympic games (1914); success in athletics (1919); w. r. seagrove, notes on athletics (1922); f. a. m. webster, throwing (1922); athletic records to date (1922); jumping (1922); c. e. hammett and c. l. lundgren, how to be an athlete (1923); j. a. mussabini, track and field athletics (1924); alec nelson, practical athletics and how to train (1924); j. f. rogers, athletics for women (1924); s. c. staley, games, contests and relays (1924); a. b. wegener, track and field athletics (1924); f. a. m. webster, athletics (1925). (f. a. m. w.) atlanta, ga., u.s.a. (see 2.853), the commercial capital of the southeast, had a population of 200,616 in 1920, an increase of 29:6% over 1910, of whom 62,796 were negroes and 4,738 foreign born; in 1925 (local estimate) it was 253,783. its area was increased from 17-3 sq. m. in 1910 to 31°6 sq. mm. in 1924. 257 the value of its manufactured products was $33,083,000 in 1909; $113,092,000 in 1919; $122,284,262 in 1923. leading industries within the corporate limits (1923) were printing and publishing, and the making of confectionery and ice cream, mattresses and bed springs, furniture, lumber and planing-mill products. cot- ton-mills, fertiliser plants and numerous other industries were located near the city. hydro electric power to the amount of 236,936,695 kilowatt hours was used in 1923. a steam heating plant, serving the central part of the business district through underground mains, evaporated 315,000,000 |b. of water in 1924. atlanta has easily kept its position as the principal distributing point and financial centre of the southeast. the seat of a federal reserve bank, it ranked 16th in bank clearings in 1924 ($2,895,571,045) and 21st in postal receipts ($3,408,368). in 1925 an average of 400 cars of merchandise and package freight jeft the city daily. by 1925 large hotels and sky-scraper office buildings gave it a metropolitan appearance. the streets were overcrowded, and projects for widening some of them were under consideration. a city-planning commission, established in 1920, secured the adoption of a zoning ordinance in 1922. in 1921-2 a comprchensive survey of the public-school system was made, and plans were mapped out to meet anticipated needs as far ahead as 1940. emory university, founded in 1914, and ogle- thorpe university, originally located at midway, but destroyed during the civil war, and re-opened at atlanta in 1916, are notable additions to the surrounding circle of educational institutions. on stone mountain, 16 m. east of atlanta, a magnificent memorial to the southern confederacy was under construction in 1926. when completed a procession of over 700 gigantic figures carved in the bare granite will sweep across the precipitous mountain side for 1,300 feet. under the central group will be quarried out a vast memorial hall, to serve as a depositary for records and relics of the confederacy.",
    "source_url": "https://archive.org/details/encyclopaedia-britannica-encyclopaedia-britannica.-3-encyclopaedia-britannica-inc.-1926",
    "observed_at": "2026-05-17 11:59:27",
    "integrity": {
        "hash_check": "match",
        "hash_scope": "full_normalized_text",
        "computed_sha256": "39b234dad9b455cce85643cdcf406bf55f59d363770c54fc25ad196019e894a0"
    },
    "machine_use": {
        "read": true,
        "cite": true,
        "decision": "verified_public_domain_text"
    },
    "documentation": {
        "white_paper_url": "https://www.goguides.com/white-paper.php",
        "pdf_url": "https://www.goguides.com/whitepapers/goguides-ai-source-clearance-white-paper.pdf"
    }
}