No. 1, May, 1921] GENETICS 51 



cent bronze. — The last cross shows also that the Narragansett pattern is probably allelomor- 

 phic to bronze. Narragansett is also likely allelomorphic to bourbon red. Reciprocal 

 crosses gave Fi much like the Narragansett but with subterminal black bands less intense 

 and slaty regions slightly auburn. — These 4 patterns evidently form a system of quadruple 

 allelomorphs. — White is recessive to color. A white 9 by a bourbon red cT gave Fi all bronze- 

 red intermediate; she evidently carried bronze but lacked the factor for color. Bronze 9 

 by white & gave bronze. F2 was 75 per cent bronze and 25 per cent white. — W. R. B. 

 Robertson. 



328. Robertson, W. R. B. Unusual tetrads and their bearing on the problem of crossing- 

 over. [Abstract.] Anat. Rec. 20: 199. 1921.— Among the ring-like tetrads resulting from 

 the pairing of compound chromosomes, such as occur in Chorthippus curtipennis, there have 

 been found again cases of a condition in which the 2 strands of 1 of the members of a pair 

 show 1 complete revolution about each other which is not present in the strands of the 

 other member of the pair. This torsion occurs, of course, in a region of the tetrad where 

 disjunction has taken place — that is, at an internode between 2 points of the tetrad which are 

 still in conjunction. — This may mean: (1) That the pairing chromosomes were each split and 

 the halves independently twisted about each other before parasynapsis took place; or (2) 

 that crossing-over between 1 strand of each of the conjugants has taken place at some pre- 

 vious time. — If the latter be the correct interpretation, then opposite sides of the ring would 

 each receive 1 strand of the paternal and one of the maternal pair, and the 1st division be 

 accordingly equational for the bulk of the tetrad. — The important point, however, is that 

 the crossing-over hypothesis gives a very satisfactory explanation of the abnormality. — W. R. 

 B. Robertson. 



329. Rtx, Georg von. Methoden einer exakten Priifung des Fortschrittes bei der Zucker- 

 riibenzucht. Paritiits- und doppelte Standard-methode. [Methods of exact testing the ad- 

 vancement in sugar-beet breeding. Parity and double standard method.] Zeitschr. Pflanzen- 

 ziicht. 7: 227-2.37. 1920. — Author discusses need for constant standard by which to judge 

 results of progress in sugar beet breeding, explaining that one lot of seed cannot be kept for 

 that purpose because of its deterioration in storage. He explains two standards and how to 

 secure them: (1) Parity method, or method of direct comparison with standard beet speci- 

 mens, in which a line is bred pure and maintained pure for purposes of comparison. (2) 

 Double standard method, or method of comparison with a corrected and doubly checked 

 standard, in which half the seed from each season is planted in turn the next two seasons. — 

 H. B. Tukey. 



330. Sakamura, T. Experimentelle Studien iiber die Zell- und Kernteilung mit beson- 

 derer Riicksicht auf Form, Grosse und Zahl der Chromosomen. [Experimental studies on cell 

 and nuclear division with special reference to form, size, and number of chromosomes.] Jour. 

 Coll. Sci. Imp. Univ. Tokyo 39^: 1-221. 7 pi., £4 fig. 1920.— The paper is chiefly of interest 

 to cytologists but the following conclusions are important for geneticists. Chromosomes 

 were found normally to be constant in number and to retain their identity. Constrictions 

 produced by attachment of spindle fibers were found to be of wide occurrence in plants and 

 animals and to serve as excellent marks of identification of particular chromosomes. Frag- 

 mentation sometimes occurs at constrictions and increases the chromosome number. Such 

 fragmentations become heritably fixed and give rise to varieties with deviating chromosome 

 numbers. Such abnormalities, as well as those due to failure of particular pairs to segregate, 

 are brought about by conditions in surrounding cytoplasm and are the effects of outer or inner 

 factors which may also produce mutation by direct action on the germ-plasm. Nearly related 

 species and varieties of some genera, e.g., Triticum, differ by multiples of the x or reduced 

 number characteristic of the ancestral species. Fragmentation, failure to segregate (non- 

 disjunction), inclusion of two or more nuclei in one cell or one membrane may be produced 

 by physical or chemical means or by influence of parasites. Author did not secure normal 

 development of pollen grains with such unusual chromosome equipment. — Leonas L. 

 Burlingame. 



