112 GENETICS [Bot. Absts., Vol. IV, 



pairs in complexes of different individuals is noted. In the cells studied, the chromosomes 

 are seldom actually associated in pairs. Results support theory of individuality of chromo- 

 somes, and in particular harmonize with the view that in the somatic cells the chromosomes 

 exist in biparental pairs. — Bertram G. Smith. 



694. Parmexter, Charles L. The chromosomes of parthenogenetic frogs. Jour. Gen. 

 Physiol. 2 : 205-206. Jan. 20, 1920. — Spermatogonial chromosome groups in frogs developed 

 from artificially parthenogenetic eggs are clearly diploid, probable number 26. Tetrads of 

 normal form are produced in haploid number. Author suggests diploid number of chromo- 

 somes is due to retention of second polar body, or to premature division of chromosomes 

 before first cleavage of egg, but work on this point is still in progress. — .4.. Franklin Shull. 



695. Payne, F. Selection for increased and decreased bristle number in the mutant strain 

 'reduced.' Anat. Rec. 17:335-336. Jan. 20, 1920.— [Author's abstract of paper read before 

 American Society of Zoologists, St. Louis, December 30, 1919.] — In the seventh generation of 

 the strain selected for extra bristle number appeared a male with 1 bristle on the^scutelluin. 

 When out-crossed to wild this reduced condition of the bristle number behaved as a sex- 

 linked character. The number of bristles in this mutant line varied from to 4. Minus 

 and plus selection lines were started and carried for 64 and 60 generations, respectively. 

 Toward the end of the experiment the percentage of zero flies in the minus line varied from 

 96 to 100. A pure line with no bristles was not produced. In the plus line the percentage of 

 flies with 4 bristles (this is the normal number) reached 64.25 in the fifty-fifth generation. 

 There were very few flies with one and no bristles. The two lines were widely different at 

 the end of the experiment. They came originally from the same pair of parents. How do 

 they differ genetically, is the question. By means of linkage it was shown that there are at 

 least two factors and probably a third concerned in the production of bristles in the plus 

 line. Two of these factors are sex-linked. If there is a third present, it is in the third chro- 

 mosome. In the minus strain only a single factor is present. — F. Payne. 



696. Payne, F., and Margaret Dexxy. Heredity of orange eye color. Anat. Rec. 17: 

 337. Jan. 20, 1920. — [Author's abstract of paper read before American Society of Zoologists, 

 St. Louis, December 30, 1919.] — Orange appeared in the mutant strain "reduced." When 

 mated to wild, three kinds of males appeared in F 2 . They were red, orange, and a new one 

 called "salmon." From this it seemed that orange was due to two factors, and that they 

 crossed over in Fi. One of these factors by itself produced salmon, the other produced no 

 visible effects. We shall call this second factor "salmon modifier." Experiments were 

 devised to test this hypothesis. If crossing-over occurs between salmon and salmon modi- 

 fier in Fi, then some of the F 2 red males should carry salmon modifier. These were tested by 

 mating to pure-line salmon females. If crossing-over occurs in Fi of this cross, salmon and 

 salmon modifier should occasionally get into the same chromosome, and hence orange males 

 should appear in F : . They do occur, and hence the assumption is justified. — The above out- 

 lines the behavior of orange when mated to wild. When mated to reduced, however, no 

 crossing-over occurs in Fi, and hence only orange and red males appear in F2. Presumably a 

 non-crossover factor is present in the reduced strain which prevents crossing over between 

 salmon and salmon modifier. This has not been sufficiently analyzed to report further. — 

 F. Payne and Margaret Denny. 



697. Pearl, Raymond, axd John Rice Mixer. Variation of Ayrshire cows in the quantity 

 and fat content of their milk. Jour. Agric. Res. 17:285-322. 6 fig. 1919. — A biometrical 

 analysis of variation in quantity per unit of time, and in quality as indicated by fat per- 

 centage, of the milk of nearly 7000 Scottish Ayrshire cows for the years 190S and 1909. In 

 this analysis the milk or butter-fat percentage for each year of age was analyzed separately. 

 — Mean weekly yield of milk with advancing age is found to be represented by a logarithmic 

 function, and to be in accordance u Lth a law which may be stated in this way: The absolute 

 amount of milk produced per unit of time increases with the age of the cow until the maxi- 

 mum is reached, but the rale of increase diminishes with advancing age until the absolute 



