NATURAL HISTORY OF THE DIATOMACE/E. 453 



detection becomes in consequence more casual and difficult." The 

 modes of conjugation have been reduced to four classes, thus: 1st. We 

 have two parent frustules and two sporangia, as the result of their con- 

 jugation. This mode has been seen to take place in the genera Epith- 

 emia, Cocconcma, Gomphonema, Encyoucma, and Colletonema. 2d. From 

 the conjugation of two parent frustules we have formed a single sporan- 

 gium. This mode has been only seen to take place in Himantidium, but 

 most likely will be found hereafter to be natural in allied forms. 3d. 

 The valves of a single frustule separate, the contents set free, rapidly 

 increase in bulk, and finally become condensed into a single sporangium. 

 This has been seen in Cocconeis, Cyclotella, Melosira, Orthosira, and 

 Schizonema. 4th. From a single frustule, as in the last mode, two spo- 

 rangia are produced in the process of conjugation. This takes place in 

 Achnanthidium and Rkabdonema. Thus far has observation gone ; but 

 no one seems to have traced the further history of the sporangium. For 

 we find no record of the undoubted production of gonidia, as they are 

 called, or seeds possessed of motion, from the contents of the sporan- 

 gium. In fact, we have no proof that such contents are developed into 

 spores, still or motile. It is true that Rabenhorst, a German observer, 

 has figured and described what he supposes to be the development of 

 gonidia or motile spores from the contents of a sporangium in a filament 

 of Melosira varians, an extremely common species found growing in 

 fresh water in all quarters of the globe. It would seem strange, there- 

 fore, that others have not seen the same thing ; and later observers have 

 doubted his record for this reason alone, apparently, that they have not 

 chanced to see it. However, Smith, one of the best of the English 

 authorities on this subject, says, "On the whole, the facts at present 

 within our knowledge seem fully to warrant the conclusions that the 

 conjugated state of the diatomaceae is the first step in the reproduction 

 process of these organisms ; and that the sporangial products of this 

 condition become the parents of numerous young frustules destined to 

 renew the cycle of phenomena which accompanies the life and growth of 

 the species from which the sporangia have thence originated." It is very 

 likely that the contents of the sporangium are converted into spores or 

 gonidia, as Rabenhorst has stated, and that after escaping and moving 

 rapidly about, and thus aiding in distributing the species, these gonidia 



