September 20, 19 17] 



NATURE 



53 



merits flying into space with accelerating speed. The 

 dissipating force, as indicated by the great prominence, 

 lies at the surface of the sun, and may be localised in 

 a very restricted area. The main stem consisted of a 

 stream of rapidly moving gas, which was biilliantly 

 luminous when it formed a continuous column, but so 

 soon as the continuity was broken by the stoppage of 

 the supply of gas from the chromosphere, the separate 

 detached masses faded very rapidly. The rapid fading 

 is probably to be explained by the extremely low 

 density of the gas involved. Mr. Evershed argues that 

 the density is so small that the gas can have no tern- 

 perature in the ordinary sense ; its emissive power will 

 thus be dependent only on absorption of photospheric 

 radiation, which is apparently insufficient to maintain 

 luminosity at great heights. A remarkable feature of 

 the great eruption was the practically simultaneous 

 fading of the entire prominence. 



Colours of Stars in Galactic Clouds. — In con- 

 tinuation of his work on the colours and magnitudes of 

 stars in clusters. Dr. Harlow Shapley has determined 

 the colours and magnitudes of 300 stars in the galactic 

 clouds surrounding the cluster Messier 13 {Astro- 

 physical Journal, vol. xlvi., p. 64). A wide range of 

 colour is apparent among these stars, and the distri- 

 bution of spectral types among the 14th magnitude 

 stars appears to be much the same in this distant 

 galactic region as in the immediate vicinity of the sun. 

 Stars of all colours are included in each interval of 

 magnitude, and so far as colour is an index of in- 

 trinsic luminosity, this may be accepted as an indica- 

 tion of considerable difference in the distances of such 

 stars. The wide dispersion in magnitude of both blue 

 and red stars suggests that the extent of the stellar 

 clouds in the line of sight is relatively very great, 

 possibly greater than the distance to the nearer 

 boundary. The cluster Messier 11 proves to be a 

 physical group in the midst of the star-clouds, which 

 on their own pa'-t have the general appearance, and 

 some of the properties, of an enormous, but definitely 

 outlined, physical system. There is as yet no certain 

 evidence of the/ existence of dwarf stars either in the 

 cluster or in the galactic clouds. The cluster stars 

 are probably giants in luminosity, and the distance of 

 the group is of the order of 15,000 light-years. 



GERM-CELLS AND BODY IN 

 INHERITANCE. 

 T N Nature for March 15 of this year (pp. 55-56) some 

 -*• account was given of a summary of Dr. Raymond 

 Pearl's researches on the progeny of alcoholised fowls. 

 A later and much fuller description of this important 

 work has now appeared in the Journal of Experimental 

 Zoology (vol. xxii., 1917, pp. 125-86, 241-310), under 

 the title of "The Experimental Modification of Germ- 

 cells." This pap>er is divided into three sections, the 

 first of which describes the general plan of the experi- 

 ments, and the second the effect upon the domestic 

 fowl of the daily inhalation of ethyl alcohol and other 

 substances, while the third discusses the effect of 

 parental alcoholism and certain other drug intoxica- 

 tions on the progeny. The general results of the ex- 

 periments have already appeared in Nature (loc. cit.). 

 Dr. Pearl alcoholised his fowls by inhalation because 

 the birds refused to drink alcohol, even if highly 

 diluted ; Prof. Stockard had previously found it impos- 

 sible to administer alcohol to guinea-pigs satisfactorily 

 by the stomach, and had therefore also adopted the 

 inhalation method. While the progeny of Stockard 's 

 .eruinea-pigs had been as a rule weakly and deformed, 

 the offspring of Pearl's treated fowls were stronger, 

 though less numerous, than those of his "controls." 



NO. 2499, VOL. 100] 



In the case of the birds the effect of the alcohol on the 

 germ-cells seems therefore to have been selective^ 

 whereas with the rodents it was utterly deleterious. A 

 possible cause of the difference, which does not seem 

 to have occurred to Dr. Pearl, may be the great con- 

 trast between the respiratory mechanism in birds and 

 in mammals ; the residual air in the lungs of the latter 

 might be expected to increase the effect of the inhaled 

 poison. Further, the excessive degradation of the off- 

 spring of Stockard 's guinea-pigs suggests that the 

 I germ-cells of those animals are peculiarly sensitive to 



adverse influences. 

 I The temptation to argue from these divergent results 

 I to the terribly practical problem of alcoholism in the 

 I human race is great, and Dr. Pearl does not altogether 

 resist it. Clearly, however, the effect of the inhalation 

 i of ethyl alcohol by a Plymouth Rock hen, or even by 

 I a guinea-pig, cannot be closely compared with the 

 { effect of alcohol swallowed by the whisky- or beer- 

 drinker. The latter effect can be studied elsewhere 

 than in biological laboratories. 

 ; Another aspect of the affection of germ-cells is 

 j illustrated for plants by Mr. S. Ikeno's "Studies on 

 the Hybrids of. Capsicum annuum," part ii., "On 

 I Some Variegated Races," in a recent number of 

 t the Journal of Genetics (vol. vi., No. 3). A variegated 

 I race of this species appeared in 1913 by mutation, pro- 

 ducing, exclusively by self-fertilisation, plants which 

 have always variegated foliage, but which differ widely 

 in the intensity of the variegation. Self-fertilised 

 flowers on green branches of a variegated plant yield 

 variegateds, in the majority of which the variegation 

 is slight. By hybridising variegated with green the 

 degree of variegation in the offspring is diminished. 

 Variegation is transmissible in either the male or the 

 female line, but the transmission " is not through the 

 nucleus, but through the cytoplasm ; especially the 

 plastids contained therein may be regarded as organs 

 of transmission," and the author believes that some 

 cytoplasm containing plastids may be introduced by 

 the male gamete into the zygote. Analogous cases of 

 plant-inheritance have been previously discussed by 

 Correns, Gregon,' and others. Variegation depends 

 upon the presence of plastids which have no power 

 of forming chlorophyll, which may, indeed, be re- 

 garded as diseased, so that though the character is 

 due to a kind of infection suffered by the germ-cells, 

 it is not strictly blastogenlc. 



The same part of the Journal of Genetics contains 

 a paper by Dr. R. Ruggles Gates on " Vegetative 

 Segregation In a Hybrid Race of CEnothera (CE. 

 rubricalyx X biennis," In which somewhat similar ques- 

 tions are raised. The bud-colour character shows 

 Mendelian segregation, which may reasonably be con- 

 sidered dependent on normal chromosome distribution 

 In meiosls. But In the size of petal there is a range of 

 variation that suggests " somatic variation and segre- 

 gation . . . determined by diversities appearing in 

 nuclear or cytoplasmic material during somatic 

 mitoses." Here, therefore, we have another example 

 of the necessity for clearing issues in the study of 

 Inheritance. 



The broader aspects of evolution are discussed by 

 Dr. Raymond Pearl in an article entitled "The .Selec- 

 tion Problem" (American Naturalist, vol. li., 1917, 

 pp. 65-91). Insisting on the necessity of experimental 

 proof and the determinative action of germinal char- 

 acters, he concludes that "natural selection Is no longer 

 i generally regarded as the primary, or perhaps even a 

 major, factor in evolution." Yet, in stating that 

 I " natural selection is, from the point of view of modern 

 genetics, a somatic theory," he surely goes far beyond 

 the available evidence, and seems to ignore the prin- 

 ciple that characters of selection-value must be re- 



