\ 



ORIGIN OF MYCETOCYTES IN PSEUDOCOCCUS. 283 



fortunately, the large size of the cells and plates often results in 

 their being cut in the process of sectioning. Even if the chromo- 

 some plate studied is entire and flat, the chance of an overlapping 

 of several chromosmes is very great, considering the large number 

 involved. The latter defect may often make a count unreliable, 

 but the cutting of a plate must throw it out of consideration at 

 once. Discarding therefore all counts in which any doubt as to 

 the number can possibly be entertained, the following data were 

 obtained: 



There are 6 plates containing 25 chromosomes; 3 plates con- 

 taining 30 chromosomes; and 5 plates containing 35 chromosomes. 

 These numbers become more impressive in view of the fact that 

 during this period not a single perfect plate carrying any other 

 but these three numbers was encountered (Figs. 12-17). 



Among the plates discarded because overlapping of chromo- 

 somes made their counting uncertain, there were several in which 

 doubt existed only with respect to a single chromsome, i.e., 

 whether to count it as one or two. Accordingly as these two pos- 

 sibilities were taken, these doubtful plates were found to fall 

 under one of the three types given, or differ from it by one chrom- 

 osome. In this uncertain group of plates there are 5 plates con- 

 taining 25, or i more or less chromosomes; 3 plates containing 

 30, or i more or less chromsomes; I plate containing 35 or 36 

 chromosomes. Again, no plate in which doubt as to one chrom- 

 some exists was found to approximate any number but 25, 30 

 and 35. 



How are these numbers to be explained? 



In every perfectly clear plate, it was evident that the chromo- 

 somes resemble each other very closely as to size and shape. This 

 fact coupled with the observation that the multiple numbers 

 found are all multiples of five makes it improbable that an irregu- 

 lar process of fractionation of the chromosomes is to be held 

 accountable. Again, although a definite regularity obtains in 

 these chromsome numbers, it has also been observed that one 

 egg may carry cells of more than one type. Thus a plate with 

 25 may be found close to another plate with 35 chromosomes. 

 This too would be difficult to explain on the basis of a fractiona- 

 tion, especially as the character of the numbers would make 



