INTRODUCTION TO CRYPTOGAMIC BOTANY. 109 



they are so evidently connected. Besides, the zoospores occa- 

 sionally arise from impregnation, as in Q^dogoniwm and Bui- 

 hoclicete. The same objections apj^ly to the zoosporous and 

 aplosporous divisions of Decaisne. 



90. There is something attractive in the main divisions pro- 

 posed by Kiitzing of Isoearpece and HeteTocavpece, but strictly 

 speaking, the tetraspores of the rhodosperms are not fruit at 

 all ; they are only a sort of buds, and in consequence, the 

 plants which bear them are more luxuriant than those which 

 bear the true fruit. Neither is the fruit always of one kind 

 only in the other division. Stilophora, for instance, with seve- 

 ral other allied Algte, bears two distinct forms of fruit, each con- 

 taining zoospores, both of which reproduce the species. 

 Bangia, moreover, and Porphyra ought by no means to be 

 in distinct main divisions, while the red-spored species of 

 Hwmatococcus pass gradually into Bangia. Besides which, 

 the seaweeds (Fucoidece) are scarcely more nearly allied to 

 Conferva than they are to Chroolepus. If, however, such 

 names as Ghlorospermeoi and Rliodospcrnfiece were substituted, 

 we should, Avith a few modifications, have undoubtedly two 

 tenable groups, and the first of these, when divided into two, 

 would make two equivalent with the first and third of Harvey, 

 As regards the subordinate divisions, they are far too nu- 

 merous, and separate plants closely allied to each other. I 

 see no advantage, therefore, whatever in adopting Kiitzing's 

 rather than Harvey's arrangement. The arrangement of Fries 

 into F'ucaeea', Ulvacece, Diatoniacece, is far inferior. It is 

 universally admitted that the divisions of Decaisne are not 

 tenable, but perhaps no one has contributed more than that 

 excellent botanist, especially in conjunction with M, Thuret, to 

 an accurate knowledge of these plants. It is scarcely neces- 

 sary to notice that of Zanardini,* which has the disadvantage 

 of joining Fucoidece and Floridece in one group, while it 

 sej)arates nearly allied species into another, Endlicher divides 

 Algffi into three orders : 1. Confervacece, 2. Phycoidece, 

 8. Floridece, which are, as nearly as possible, synonymous with 



* Saggio di Class, delle Ficee, 4to, 1843, 



