A STUDY OF SOMATIC CHROMOSOMES. 335 



destaining. Foot and Strobell ('13, p. 199) have particularly 

 objected to such explanations, characterizing them as "the old 

 story so familiar to cytologists if a feature is where, hypo- 

 thetically, it ought not to be, it is an artifact, and if it is not 

 where it ought to be, it is due to faulty technique." Their 

 objections are valid, I think, up to a certain point. Variations 

 may easily occur, in fact the double chromosome groups in 

 Anasa are proof that they may occur frequently. As to the 

 total disappearance of chromosomes I have no proof. Moreover, 

 the metaphase plates which have been studied are all reasonably 

 clear and no variations in the number of type of the chromosomes 

 have been demonstrated, other than the above. It seems reason- 

 able, therefore, not to entirely ignore the above-mentioned expla- 

 nations, especially since the preponderance of proof is in favor 

 of the characteristic chromosome number. If an entire chromo- 

 some plate lies in one section, and is sufficiently clear to be 

 counted, the normal number has invariably been found (with 

 the exception noted). It is well to point out that the study of 

 prophases has been found to be unadvisable owing chiefly to the 

 inaccuracy of the enumerations, due to the overlapping of the 

 chromosomes. 



Figs. 1-4 show plates of the 2i-chromosome type. To the 

 left are the chromosomes as they appear in the plates, and to the 

 right the chromosomes linearly arranged. Fig. I is a spermato- 

 gonial metaphase. Fig. 2 a late cleavage metaphase, Fig. 3 is 

 from a neuroblast in division, and Fig. 4 from an abdominal 

 mesoderm cell. The last two figures are from the same embryo 

 at a stage of development corresponding to stage II. 



Bearing in mind the 21 -chromosome complex previously men- 

 tioned, it will be seen that in all these figures there are 2 m- 

 chromosomes and 3 chromosomes larger than the others. One 

 of these three larger chromosomes, the largest, is obviously 

 unpaired. In the eight pairs of mesochromosomes there is a 

 gradual gradation in size from the macrochromosomes to the 

 w-chromosomes. Aside from this decrease in size they show no 

 striking peculiarities. The attempt to arrange the chromosomes 

 in pairs is, of course, an arbitrary one, but it will readily be seen 

 that, with the exception of the unpaired largest, there are two 



