DENDROCOMETES PARADOXUS. 339 



tlie next stage there is a differentiation of the skein into a 

 number of lines, which stain very faintly in iron-hfemato- 

 xylin, but a little more deeply with iron-brazilin, and a band 

 of deeply staining rods and granules having the usual 

 chromatin reactions. At this stage the micronucleus is 

 frequently 10^ in diameter (PI. 17, figs. 4, 11, 12). The exact 

 determination of the nature of the equatorial band of chro- 

 matin granules is a matter of considerable difficulty. With 

 the highest powers of the microscope the granules or minute 

 rods appear to be connected together in the manner shown 

 in PI. 18, fig. 13, but still I do not feel so far convinced of 

 this connection as to deny the proposition that they are 

 isolated chromosomes. With lower powers of the microscope 

 they have every appearance of being a band of rod-shaped, 

 chromosomes (PI. 17, fig'. 12). Whatever the future may 

 reveal regarding these bodies^ I think it is clear that the 

 chromosome elements are numerous, — too numerous, in fact, 

 to count with any degree of accuracy. 



The micronucleus next becomes somewhat oval in shape 

 (PI. 18, fig. 16), and the band divides into two bands. The 

 faintly staining lines, which we may call the linin threads, 

 are arranged in a roughly parallel manner forming a some- 

 what tub-shaped spindle. They do not come to a point at 

 each end of the figure, and there is never any dot or granule 

 that can suggest the presence of a centrosome. In the n-ext 

 stage (PI. 18, fig. 17) the two bauds of chromosomes have 

 passed to the extremities of the figure, and soon become 

 aggregated together to form a single irregular lump. In 

 some cases there is a clear vacuole present at one of the 

 poles (the right pole in PI. 18, fig. 17), but I have seen it so 

 rarely that I am at a loss to understand its meaning. In 

 the next stage (PI. 18, fig. 18) the spindle becomes very 

 much more elongated, so that the whole figure may be 

 25 — 30 /x in length; the chromatin is in the form of a single 

 spherical lump surrounded by a clear zone at the points o 

 the spindle. The spindle next becomes detached from the 

 new micronuclei (PI. 17, fig. G), and gradually dissolves in the 



