SOIL PROTOZOA. 327 



formation of irregular chromosomes, some of which appear rod- 

 like and others rather rounded. While these changes are taking 

 place the whole nucleus increases in size and becomes barrel- 

 shaped, attaining a length of 6-7 fi. A few spindle-fibres make 

 their appearance within the nucleus, but no definite spindle 

 comparable with that found in Amoeba lawesiana is produced. 

 Moreover, the fibres seem to lie on the nuclear membrane rather 

 than within the cavity of the nucleus. The chromosomes now 

 become arranged on the fibres, but I have not discovered any 

 examples which show all the chromosomes arranged in an 

 equatorial ring or band. All the stages of this phase of division 

 show two principal groups of chromatin granules or chromosomes 

 at each end of the long axis and other chromosomes irregularly 

 disposed in the equatorial region. In the latter region the 

 individual chromosomes are exti'emely difiicult to distinguish, 

 and it is therefore practicallj'' impossible to count them. Thei'e 

 appear, however, to be about eight chromosomes or chromatin 

 masses produced within each nucleus, four of which travel to 

 each end. A description of these stages of division is very 

 difiicult to make owing to the fact that no two stages exactly 

 agree, as will be seen on referring to figs. 68-70. All of them 

 ai-e, however, of the same general type, and a detailed description 

 of each is unnecessary. The chi'omosomes ultimately become 

 drawn to the two poles of the nucleus, a stage which is well 

 shown in fig. 71. 



Shortly after this, or even earlier, the chromosomes begin to 

 break up and lose their distinct outline. As a result of this, 

 a granular mass of chromatin, rather triangular in outline, is 

 produced at each end of the nucleus (figs. 72 <k 73). The nucleus 

 now begins to elons^ate, and the chromatin is reoi'ganised into 

 daughter-nuclei, which thus gradually sepai^ate further and 

 further apart. In fig. 73 a stage is shown in which a dumb-bell 

 appearaiice is presented by the two rounded daughter-nuclei and 

 the nuclear membrane consti'icted between them. This is the 

 latest stage of division that I have obtained. I have failed to 

 discover any stages showing fission of the animal, and it seen?s 

 to be fairly evident that this occui's after nuclear division is 

 quite complete. In this connection it is interesting to note that 

 I have found a large number of bi-nucleate amoebae on the same 

 preparation. It is possible that these are forms in which fission 

 is retarded, or again they may be abnormal individuals, for I 

 have found one or two tri-nucleate forms as well. 



(c) Remarhs. 



The nuclear division in Amceha agricola differs from that which 

 occurs in any other amoeba. It is obviously a modified mitosis, 

 but it is not easy to connect it up with any of the numerous 

 mitotic nuclear divisions which have been described and figured 

 in other amcebse. 



22* 



