FRESH-WATER ALG^ OF THE UNITED STATES. 189 



two to five, more or less tumid, single in each sporangium, at maturity reddish or yellowisli fuscou.?, 

 before germination dividing themselves into (mostly four) zoospores. 



Antheridia shortly filiform, 1-2-3-10 articulate, mostly single, either upon the si)orangium or 

 Tegctation cell. 



Remarhs. — The QiJdotjoniacrw have been by previous writers simply divided into 

 two genera, (Edoijonium and Bulhochcvte. The plants represented by these two 

 divisions have certainly many characters in common, as in the production of their 

 zoospores and spermatozoids as well as in tlieir peculiar method of cell division. 

 Yet they are so very diverse in some particulars in regard to the latter, as well as 

 in their habit of growth and in the formation of their sporangia, that it has 

 seemed to me that the diiferences between them were more tlian sufficient to clia- 

 racterize merely genera, and that to each of these groups should be awarded the 

 rank of a svib-family. 



Again, in the old genus of CEdogonhtm, we have very distinct groups, separated 

 by differences in the most important of all the cliaractcristic portions of the plant — 

 the sexual apparatus. These groups are the so-called Moncecious, Gynandrous, and 

 Dicecious (Edogonia; the moncEcious division comprising those plants in which one 

 individual gives origin both to the female and male germs; the gi/nandwus, those 

 species in which the plant that produces the female germ gives origin also to a 

 peculiar zoospore, the so-called androspore, which, after a period of motile life, 

 settles down and develops a dwarf plant, the androecium, in which the sjici-mato- 

 zolds are developed ; and the dicecious group containing species in which the male 

 and female plants are distinct individuals. Dr. Pringsheim states (MorjJtoIogie der 

 OEdofjon., p. 43) that these groups pass into one another, but in my opinion, by his 

 own showing, they are sharply distinct. The nearest approach to such passage is 

 between the first and second groups, and consists simply in the fact tliat in certain 

 species the androspore when it settles down develops into a one-celled instead of a 

 two or three-celled anthcridium. This to me does not seem to indicate a union of 

 the groups, for the essential difference is not in the form or complexity of the an- 

 thcridium, but in the circtnnstance that in the one case the female filament develops 

 a spermatozoid capable of fertilizing the germ, whilst in the other it gives rise to 

 a body which does not possess that power at all, but does have tlie capability of 

 giving origin to a second plant, in which the spermatozoid is developed. The 

 groups, therefore, appear to be sharply and distinctly definable. 



In the Bidhoclia'tiiC but a single genus has as yet been discovered, and this is 

 distinctly gynandrous, but it seems probable that hereafter other plants of this 

 subfamily will be found which are monaecious or disecious, so that we will have 

 in the two subfamilies two parallel groups of genera. 



For the reasons above indicated I have ventured to divide the family into two 

 subfamilies, the one comprising three, the other a single genus. The pcctdiarities 

 of growth, production of zoospores, and sexual development will be found described 

 under the particular subfamilies. 



