594 Transactions of the Society. 



is probably to be explained in this way. When the cultivation 

 was removed from the incubator, some bacilli were growing by 

 fission, some were forming spores, and some had passed into a state 

 ready to form spores. The first go on growing by fission, the last 

 complete their spore-formation, which was arrested by removal 

 from the warm temperature. That actively growing rods would 

 not have formed spores so early is evidenced by the facts observed 

 in the second series of observations on the formation of spores. 

 The next thing that is observed is that several of the spores take 

 on the stain, and are as intensely violet as the adult bacilli (see 

 fig. 7), The number of the spores which take on the stain in 

 this way goes on increasing as time passes, till in about four hours 

 almost all the spores stain violet. In three hours the first indica- 

 cation of sprouting of these spores becomes evident. The stained 

 part of the spore loses its oval shape, becomes elongated, and ia 

 soon seen to burst through the spore-capsule at one part (see 

 fig. 7). It then presents the appearance of a short rod, with a 

 pale envelope embracing one end. This rod gradually leaves the 

 spore-capsule and then goes on multiplying as a full-grown bacillus. 

 In specimens taken from four to five hours all stages of growth can 

 be seen, and the remains of the ruptured spore-capsules are evident 

 (see fig. 7). 



The bacilli appear to grow mainly by fission, but I have seen 

 appearances which seem to me only explicable on the supposition 

 that they also grow by sending out buds from one end (see fig. 11). 

 A bacillus may be seen with a small somewhat conical stained point 

 attached to one end, though separated by a marked division. This 

 is certainly not the common mode of growth by fission, for there 

 the rod seems to divide into two pretty equal halves, while here 

 we but have a minute piece attached to one end. 



The mode of formation of spores may be traced in a similar 

 manner to that described above in the case of the sprouting of the 

 spores. It is, however, as a rule necessary to leave the organisms 

 to grow for a much longer time than in the former instance. I 

 have not found development of spores as a rule before twenty-three 

 hours, but this depends very much apparently on the amount of 

 fluid that was present and the number of bacilli introduced at the 

 time of inoculation. The first thing noticeable is that the rod 

 begins to swell and becomes spindle-shaped (see fig. 8). This swell- 

 ing, which generally affects the middle of the rod, but may in some 

 cases be most marked toward one end, increases in size, and the 

 centre of the swelling gradually ceases to take on the stain (fig. 8). 

 The capsule of the spore apparently is also formed within the rod, and 

 is not merely the outer part of the rod. In three or four hours the 

 rod is seen to have almost or completely disappeared, leaving the 

 spore lying free or within the faint outline of the original bacillus 



