SOIL PROTOZOA. 317 



division. For example, fig. 28 shows an organism in which the 

 flagella are widely separated, although the nucleus shows no sign 

 of approaching division. I am unable to state the fate of the pair 

 of rhizoplasts. 



In the earliest stages in the division of the nucleus the 

 chromatin seems to undergo some process of dissolution and re- 

 organisation whereby certain parts of it, which stain less intensely 

 than the rest, gradually come towards the centre of the nucleus. 

 These take the form of roundish or irregular granules which, 

 at the stage shown in fig. 31, appear to be arranged in a fairly 

 regular manner on a kind of leticulum or network of linin 

 threads. Whilst these changes are taking place, the remainder 

 of the chromatin, having deeply staining properties, becomes 

 arranged in the form of a ring round the periphery of the 

 nucleus. This does not always happen in the very early stages, 

 however, as is shown in fig. 33, where the deeply staining 

 chromatin is still present as tliree blocks, whilst at the centre of 

 the nucleus there is a group of lighter granules. The changes 

 which occur in the lighter staining gi'anules are very diificiilt to 

 make out, but it seems as though they gradually concentrate 

 towards the centre of the nucleus, and there become arranged in 

 an irregular manner on a kind of plate. In figs. 32 and 33 there 

 appear to be six principal granules disposed in two bands upon 

 what seems to be a spindle formation. I do not Avish, however, 

 to lay stress on the pi-esence of a spindle within the nucleus, for 

 I have failed to make it out with any degree of distinctness, and 

 even in those examples which present the spindle appearance, 

 there are always irregularly-branching linin strands running in 

 various directions, as shown in figs. 35 and 36. 



In the stages represented in figs. 34-36, four principal roimd 

 granules are present. These represent the nearest approach to 

 chromosome formation in the M'hole series of changes. It would 

 appear from these figures that the nucleus produces division 

 centres fi^om within, and is not dependent on the migration 

 of the blepharoplasts to their antero-lateral positions for the 

 formation of its poles of division. I have not been able to trace 

 further the movements of the four internal chromosomes. In 

 the succeeding stages the nucleus elongates somewhat laterally, 

 and the peripheral ring of chromatin begins to break up and 

 travel towards the latei'al poles, at the same time advancing on 

 to the dorsal and ventral surfaces of the nuclear membrane. In 

 fig. 38 the connections between the blepharoplasts and the 

 nucleus which ultimately become the rhizoplasts are fairly 

 clearly seen. The lateral elongation of the nucleus now becomes 

 moie pronounced, and the chromatin, travelling along the linin 

 threads of the nuclear network, becomes arranged in small 

 gi^anular masses towards the lateral poles. At this time the 

 threads stretching across the centre of the nucleus can be made 

 out fairly easily (figs. 39 & 40). These linin threads are doubt- 

 less absorbed, and two laterally situated daughter-nuclei are thus 



