March 8, 1917] 



NATURE 



25 



and more water the jelly sets more firmly and cuts 

 more easily, but is less transparent. 



I have used such jelly in preparing sections of the 

 wings of insects. The wing was first placed in an 

 ordinary silvering solution containing Rochelle salts 

 to which 50 per cent, of alcohol had been added. 

 After it had thereby become blackened from the de- 

 posit of metallic silver it was washed with 50 per 

 cent, alcohol and then placed in rectified spirit. To 

 embed the wing after this treatment I formed a cell 

 of plasticine on a sheet of glass. A layer of melted 

 jelly was placed in the cell, which was then filled up with 

 alcohol. The wing was placed in the cell. It dropped 

 through the alcohol on to the surface of the jelly. The 

 alcohol was at once run off by making a cut through 

 the walls of the cell. The latter was then filled up with 

 more of the jelly. By this procedure wings of moths 

 iind butterflies, which are not readily wetted by water, 

 could be obtained firmly embedded, free of air-bubbles 

 and without displacement of the scales. To the 60 c.c. 

 of water used in making the jelly I had added 16 

 grams of hyposulphite of soda. Having cut the 

 lump of jelly containing the wing into ten slices of 

 ■equal thickness with a Gillette razor-blade, these 

 slices were threaded in order on a wire, and placed 

 all night in a half-saturated solution of tartaric acid 

 in 70 per cent, glycerine. The acid decomposed the 

 hypo., liberating sulphur, with the result that the jelly 

 acquired an ivory-white colour, on which the wing- 

 sections appeared in black. The slices were mounted 

 'H a cell containing glycerine jelly. Embedding in 



Iluloid no doubt would be preferable for wings of 

 mailer insects. The method here described is prob- 

 ably more suitable for larger insects, the wings of 

 which would be likelv to offer difficulties in an attempt 

 > cut thin sections. E. H. Hankin. 



Agra, India. 



National Service. 



The wording of the enrolment form for National 

 Service having given many the impression that volun- 

 teers are wanted onlv for industrial work, which some 

 men over fifty cannot possibly undertake, I wrote to 

 the Director of National Service for definite informa- 

 tion, and my queries were answered as follows : — 



"National voluqteers are required not only for in- 

 dustrial work, but also for other positions of national 

 importance." 



Further : " Mr. Chamberlain wishes it to be clearly 

 understood that brains as well as ' hands ' are required, 

 and also that no volunteer will be set to do work for 

 which he is not fitted personally." 



These very clear and authoritative replies will, per- 

 haps, relieve the doubts of many who have been 

 hesitating on the very reasonable ground that the 

 work involved seemed to be bevond their capacity. 



C. Welborne Piper. 



THE CLASSIFICATION OF HELIUM STARS. 

 \^7HILE it is now generally admitted that the 

 * * spectroscopic differences between the 

 different classes of stars are mainly due to differ- 

 ences of temperature, there are two widely diver- 

 gent views as to the order of celestial evolution 

 which may be inferred. In one of them, the 

 evolution is supposed to proceed by a continuous 

 decline of temperature from the white to the red 

 stars ; in the other, which has been consistently 

 advocated by Sir Norman Lockyer during nearly 

 thirty years, it is maintained that the prog^ression 



NO. 2471, VOL. 99] 



is from stars at a low stage of temperature to the 

 hottest stars, and from these to stars at a low 

 temperature, so that a given star will have the 

 same temperature twice in the course of its evolu- 

 tion. The classifrcations of Rutherfurd, Secchi, 

 and Vogel, and the expansion of these into the 

 Harvard system, may be interpreted in terms of 

 the first hypothesis, though actually they may be 

 regarded as merely empirical and independent of 

 any such consideration. The classification of 

 Lockyer, on the other hand, is essentially based 

 upon the supposition that there must be stars 

 which are getting hotter as well as stars which 

 are cooling, in accordance with the theory of con- 

 densing masses of gas or swarms of meteorites. 



If the spectrum of a star depended solelv ujjon 

 the surface temperature, there would evidently be 

 no observational means of distinguishing between 

 the two hyf>otheses. But Lockyer finds that when 

 stars at any given stage of temperature are 

 brought together by reference to the relative in- 

 tensities of certain lines, selected according to the 

 indications of laboratory experiments, they are 

 divisible into two distinct groups. The sjjectra, 

 therefore, seem to depend in part upon physical 

 conditions other than those imposed by tempera- 

 ture alone. The difference is quite probably due 

 to a difference in the degree of condensation, and 

 Lockyer 's interpretation assigns one of the groups 

 to the ascending, and the other to the descending, 

 branch of the temjjerature curve. The Harvard 

 classification takes no account of these dififerences, 

 and is accordingly along one line of temperature 

 only. 



The difference between the opposing views as 

 to the order of celestial evolution is clearly of a 

 very fundamental character, and it is important 

 that the question should be attacked in as 

 many ways as possible. The work of Prof. H. N, 

 Russell (N.ATURE, vol. xciii., p. 283) on the abso- 

 lute magnitudes of stars has already given consider- 

 able support to the main principle of Lockyer 's 

 classification, by especially emphasising the idea 

 that the order of celestial evolution is primarily 

 one of increasing density, with a maximum of 

 temperature near the middle of the sequence. As 

 regards the helium stars. Dr. Ludendorff found 

 in 191 2 (N.^TURE^ vol. Ixxxviii., p. 424) that the 

 radial velocities showed a very decided systematic 

 difference for the ascending and descending stars 

 classified by Lockyer, a difference which was not 

 so clearly shown when the velocities were referred 

 to the Harvard sub-classes. 



Since the publication of Ludendorff 's results, 

 Lockyer has supplemented his original catalogue 

 of 470 stars by a second catalogue of 354 stars, 

 and a third catalogue of 287 stars, photographed 

 and classified at the Hill Observatory, Sidmouth 

 (Hill Obs. Bull, Nos. 3 and 5). The first attempt 

 to utilise some of the additional data which have 

 thus become available has been made by Dr. 

 B. P. Herassimovitch in a recent communication 

 to the Petrc^rad Academy of Sciences (Bull. 

 Acad. Imp. Sci., 1916, p. 1419). In this paper 

 the helium stars included in the first two cata- 



