A STUDY OF SOMATIC CHROMOSOMES. 351 



('oo) reports a number varying between 15 and 19, though he 

 considers 18 as the normal number. However, he does not 

 state if this number is the haploid or diploid; the inference is 

 that it is the former. The counts of somatic chromosomes 

 reported by Farmer, Moore and Walker were made from cells 

 in a normal rectum in comparison with cancerous tissue. Wie- 

 man worked on a white human embryo. He points out the fact 

 that the number is at variance with that reported by most 

 investigators as the diploid number of the sex cells, and suggests 

 that the varying counts which he made in the soma were due 

 to a breaking up of some of the larger chromosomes. The 

 difference between the somatic and spermatogonial numbers 

 may be due, he thinks, to a similar process. Wieman also 

 suggests that a doubling of the chromosomes may have taken 

 place in Winiwarter's material. Precocious division of some of 

 the chromosomes may also give this result. 



In the somatic cells of the rabbit Winiwarter obtained a 

 chromosome number varying beween 40 and 80. The majority 

 of counts showed 42, and this he considers the normal number. 

 This number corresponds to the number of chromosomes in the 

 oogonia, but a haploid number of 10-12 in the maturation divi- 

 sions of the egg was reported by Honore, one of Winiwarter's 

 students. This latter count closely approximates the observa- 

 tions of Bachhuber on the number of chromosomes in the sperma- 

 togonia. W 7 iniwarter draws an analogy to the case of Ascaris. 

 However, the analogy will only hold if the "oogonium" was in 

 reality a follicle cell or some other non-germinal cell, otherwise 

 it would seem to be analogous to the double reduction reported 

 by Guyer in man and the domestic fowl. Apparently the cell in 

 question was not an oogonium. Flemming modifies his number 

 of 24 chromosomes by the statement that the number may be 

 even higher. Barratt's numbers are intermediate between the 

 two extremes. With reference to the varying number of chromo- 

 somes which he reports, he suggests that such variation may be 

 entirely normal and occur regularly. 



Flemming quotes Bardeleben's count of 16 chromosomes in 

 the spermatogonia of the guinea pig, but fails to find a number 

 corresponding to this in a study of an embryo. His number of 



