NATURAL HISTORY OF THE DIATOMACKffi. 45 I 



chrome of two approximated fronds [using this term instead of frustulc, 

 to indicate the perfect individual], this mixed endochromc developing 

 around itself a proper membrane, and thus becoming converted into the 

 sporangium." The sporangium is what may be called the seed vessel, as, 

 although it does not contain seeds, in the ordinary acceptation of that 

 term, yet from it proceed the new individuals who are to perpetuate the 

 species. "In a very early stage of the process, the conjugated frustules 

 have their concave surfaces [it must be remembered that we are speaking 

 of EpitJicmia, whose outline is somewhat bow-shaped, having convex and 

 concave surfaces] in nearly close opposition ; and, it may be observed, 

 that from each of these surfaces two protuberances arise, which meet two 

 similar ones in the opposite frustulc; these protuberances indicate the 

 future channels of communication by which the endochromc of the two 

 frustules becomes united, as well as the spot where is subsequently devel- 

 oped the double sporangium, or, rather, the two sporangia. The mixed 

 endochromc occurs at first as two irregular masses between the connected 

 (conjugating) frustules ; but these masses shortly become covered with a 

 smooth cylindrical membrane, the young sporangia, which gradually in- 

 crease in length, retaining nearly a cylindrical form until they far exceed 

 in dimensions the parent frustules, and, at length, when mature, become, 

 like them, transversely striated upon the surface. Around the whole 

 structure a considerable quantity of mucus has during this time been 

 developed, by which the empty frustules are held attached to the sporan- 

 gia." Thus we see that, whether any two particular frustules are drawn 

 towards each other or not (which we do not know), yet two contiguous 

 individuals pour out their respective contents, which, melting together, 

 are thus fused into a mass, around which is formed one or two new sili- 

 ceous coats exactly alike in structure, but only differing in size, being 

 larirer than those which enclose the parent frustules. Enveloping these 

 sporangia, or large cases from which the new individuals shall be evolved, 

 is thrown a protecting, or, perhaps, nutritive globular mass of transparent 

 mucous material. In different genera, slight variations are met with in 

 the method of conjugation, as described. Thus, in some species of Gom- 

 phenema, which grows attached to the end of branching stalks or stipes, 

 very much after the manner of the leaves on the ends of the twigs of a 

 tree, the sporangia lie in a direction parallel to the empty parent frustules 



