FRESII-WATEK ALU.E OF TUE U i\ 1 T E D STATES. 161 



and tliis continues until only the centres of the ends are in appof-ition, and in a 

 little while even these separate. This certainly, at least, is tlie process in certain 

 species ; but I have thought, that in other cases cells were separated by a simple 

 splitting of the end wall, each cell retaining its half of the partition. 



The zygospores are produced by a process of union of two cells, to which the 

 name of coiijugatio7i has been given. Very rarely, if ever, is there any difference 

 between the cells before conjugation, and it has not existed in any species which has 

 come under my notice; but, after conjugation, the receiving cell is frequently 

 enlarged, the other remaining cylindrical. De Bary, however, states that he has 

 found a small but constant difference between the fertile and sterile cells of 

 Spirogjjra Heeriana. 



The first perceptible change in a cell about to produce a resting spore, appears 

 to be a loosening of the primordial utricle from the outer wall, and a contraction 

 of it upon the cell contents, which thus are crowded together and more or less 

 deformed. Simultaneously with tliis, or a little after or before it, the side waU of 

 the cell is ruptured and a little pullulation or process is pushed out, which directly 

 coats itself with cellulose and rapidly enlarges to a considerable diameter, at the 

 same time growing in lengtli until it meets a similar process pushing out from an 

 opposing cell, or has attained as great a length as its laws of development will 

 allow. "When two processes meet they become fused together, the end walls are 

 ruptured, and the contents of one cell passing over are received within those of the 

 other, or else the contents of both cells meet within the connecting tube and there 

 fuse together. This is the more common mode of conjugation, in which two cells 

 of distinct filaments become joined together by a connecting tube. It is evident, 

 that, if the filaments are fertile to their fullest extent, there will be as many of these 

 connecting tubes as there are pairs of cells in the filaments, and a ladder-like body 

 will be formed, the original filaments corresponding to the side pieces, the connect- 

 ing tubes to the rounds. Hence this method of conjugation has received the name 

 of scalariform. 



In the so-called ^'■lateral conjugation" instead of cells of different filaments join- 

 ing, adjacent cells of one filament unite togetlier to complete tlie process. The 

 union of the two cells appears to take place in several ways. In accordance with 

 one plan (fig. 1 o, pi. 14), connecting tubes, puslied out from near the ends of 

 the cells, grow for a short distance nearly at right angles to the long axis of the 

 filaments, and then bend at a right angle to themselves so as to run parallel to 

 the filament-cells. The ends of these processes are, of course, opposed to one 

 another, and coming in contact fuse together so as to form a continuous tube for 

 the passage of the endochrome. Another method by which neighboring cells are 

 sometimes connected is by the formation of coadjacent pouch-like enlargements of 

 the opposing ends, and a subsequent fusion of these newly formed enlargements 

 by the absorption of the end wall between them. (See fig. 2, pi. 14.) 



Sometimes I think the union of two neighboring cells is facilitated by a curved 

 neck forming to one or both of them, so that they are bent at an angle to one 

 another, and can readily be united by means of a straight tube. 



There is still another method of conjugation, the so-called genvjlexuous, in wliicli, 



21 August, 1872. 



