324 M. A. Schneider^s Contributions to the 



is still unintelligible to me. In any case some contrivance for 

 this purpose, however simple, must exist either between the two 

 filaments, or at the side of their points of issue from the mem- 

 brane. 



Reproduction. 



During the swarming state a division of the substance of the 

 body goes on uninterruptedly, at all hours of the day. The dif- 

 ferent stages of this process follow one another with greater or 

 less rapidity in proportion as the conditions of nutrition are more 

 or less favourable. Soon after the commencement of fermenta- 

 tion in an infusion, the rate of increase attains its maximum ; it 

 then diminishes as the fermentation ceases, the offspring at the 

 same time undergoing a diminution of size. 



The commencement of the process of division is indicated by 

 the uniform distribution of the granular substance. A constric- 

 tion of the substance then takes place, usually commencing on 

 one side ; by this the body is divided into two parts, which are 

 still enclosed in the uninjured investing membrane. Simulta- 

 neously with, or perhaps before the completion of this bisection, 

 the nucleus also divides (fig. 3). Although no constriction of 

 the nucleus was ever noticed, nothing certainly was observed to 

 contradict the supposition, that the second nucleus was produced 

 in this mamner. The two halves then become constricted from 

 their surfaces of contact, in such a manner that the constriction 

 of one half crosses that of the other at right angles (fig. 4). To 

 every depression thus produced on the one side there is a corre- 

 sponding elevation of the other. The quadrisection (figs. 4 & 5) 

 then takes place suddenly as if by cutting, and without any ap- 

 pearance of a circular constriction, each portion containing its 

 proper nucleus. The divisions now acquire an oval form, and 

 arrange themselves in such a manner that the ends of the pos- 

 terior pair, which are turned towards the middle, alternate with 

 those of the anterior pair in the same place (fig. 6). In very 

 favourable circumstances (as for instance, at the commencement 

 of fermentation), a third division into eight parts takes place, 

 each division being still furnished with a nucleus. As a general 

 rule, however, the young individuals acquire filaments soon after 

 the quadrisection, and move about in various directions within 

 the investing membrane, until this bursts, and the young, which 

 are exactly like the mother except in their smaller size, are set 

 free. In favourable circumstances the empty membrane remains 

 with the two filaments. After the division of the substance into 

 four or eight parts, the investing membrane is always visible 

 without the employment of any reagents. This has not escaped 

 Jlhrenberg {loc. cit. and tab. i. xxxii.) ; he explains the appear- 



