No. 1, May, 1921] GENETICS 47 



oocyte, the second in an early stem cell of the ovary, and the third in the early cleavage of 

 the individual in which it appeared. It is possible that the orange eye arose by a mutation 

 of the W gene in a somatic cell of the early embryo. The commonly accepted view that 

 mutations are more likely to occur in gametes or germ cells near the period of maturation 

 receives no support from the data on this series. It is shown, however, that there is a much 

 greater chance that any mutation will show itself in a single individual than in several. Fur- 

 ther study of the allelomorphs of the white-eye series suggests other important conclusions 

 with respect to the origin of mutations. Since mosaic mutants involving recessive sex-linked 

 genes are always males, it would appear that mutations occur in only one member of a pair of 

 chromosomes at any one time. If the event which produces a mutation is so localized as to 

 affect a single locus in only one of a pair of homologous chromosomes it seems unlikely that the 

 artificial influencing of the kind of mutation is a possibility. It is further noted that the 

 variations of the W gene are not random deviations for they are all in a definite direction, 

 and the extreme variants seem to be the more common. For this reason selection if concerned 

 with this locus alone would not be cumulative. — H. H. Plough. 



306. MuLLER, H. J., AND E. Altenburg. a study of the character and mode of origin 

 of eighteen mutations in the X-chromosome of Drosophila. [Abstract.] Anat. Ree. 20: 213. 

 1921. — Since the 18 mutants found in the experiment of the writers on mutation frequency 

 were non-selected or random samples of (detectable) mutants in the sex-chromosome, a study 

 of them furnishes quantitative data bearing on the nature of mutations. (1) All were lethals 

 or sub-lethals. Of the 5 sub-lethals, 4 produced morphological abnormalities. (2) All were 

 completely recessive except 1 mutant of the yellow-mouse type. (3) Half of the loci involved 

 are crowded into the 1.5 units space to the left of white eye (the rest being scattered rather 

 evenly). This indicates that this region of the chromosome is really much longer than the 

 map represents. (4) All the lethals gave negative tests for "deficiency," hence deficiencies 

 are evidently much rarer than ordinary lethal mutations. (5) Three lethals were allelo- 

 morphs of known non-lethal factors, and 2 of these lethals were allelomorphs of each other. 

 Of the latter 1 became dominant in its lethal effect when crossed to anon-lethal allelomorph. 

 (6) Lethals very near 'duplicated' loci remained unaffected by the 'duplication.' (7) Muta- 

 tion occurs with not markedly different frequency in the 2 sexes, for 7 of the lethals were found 

 in the maternal, 11 in the paternal chromosome. (8) These mutations occur not only near 

 maturation, but also in earlier germ cells, in either sex, as shown by the original appearance 

 of some of the lethals in 2 sisters simultaneously. (9) Two of the original mutant indi- 

 viduals contained 2 different lethals at once; in 1 case these were in opposite chromosomes, 

 in the other case in the same chromosome. — H. J. Muller and E. Altenburg. 



307. Newman, H. H. The experimental production of twins in the starfish Patiria miniata: 

 with a discussion of the causes of twinning in general. [Abstract.] Anat. Rec. 20: 190-191. 

 Jan. 20, 1921. — A series of separate twins and of double monsters were produced under 3 

 different experimental conditions: (a) As the result of an extremely belated parthenogenetic 

 development; (b) as the result of fertilizing Patiria eggs with the sperm of another species 

 of starfish; (c) as the result of overcrowding normally fertilized eggs. All 3 methods 

 involve retardation of development, with loss of precise axiate organization at some- 

 critical period. Redifferentiation or resumption of axiate organization occurs, but unity of 

 organization has been lost, so that 2 or more axes or gradients appear instead of the original 

 1. Thus twins or double structures arise. — A series of twin types are produced which repre- 

 sent the results of differences in the earliness of onset of retardation and more or less com- 

 plete recovery. The series includes completely separated half-sized and quarter-sized blas- 

 tulae and gastrulae, full-sized gastrulae, full-sized gastrulae with 2 or more archentera, larvae 

 in which the archenteron branches anteriorly into "dicephalous" larvae, and advanced 

 bipennariae with paired, instead of only left-hand, madreporic pores and pore-canals. This, 

 physiological theory of twinning agrees with the writer's formerly expressed theory to explain, 

 the cause of specific polyembryony in the armadillo. — H. H. Newman. 



