60 MORPHOLOGY OF ANGIOSPERMS 



gus, Castanea, Corylus, and Carpinus, in the last-mentioned 

 form finding a large central mass of sporogenous tissue. Later, 

 Chamberlain 35 found that there are sometimes two or three 

 cells or even six in the archesporium of Salix, and occasionally 

 five or six in that of Populus tremuloides. Then Conrad 53 

 described the archesporium of Quercus velutina as consisting 

 of a mass of twenty to sixty or even more cells, all of which are 

 megaspores (Fig. 23). The archesporia of Casuarina, Car- 

 pinus, and Quercus are certainly not all hypodermal, like those 

 of the Rosaceae, in which the resemblance to the development 

 of the microsporangia is striking. In Juglans cordiformis 

 Karsten 64 has also found an extensive sporogenous tissue. A 

 several-celled or even a many-celled archesporium, therefore, 

 seems to be a frequently expressed tendency among the Amen- 

 tiferae,* although it is by no means uncommon among them 

 to find the archesporium consisting of a single cell, as in Alnus 

 and Betula. 



Among the Ranunculaceae great irregularity in the num- 

 ber of archesporial cells is found even in a single species. 

 Guignard 17 first found that in Clematis cirrhosa the archespo- 

 rium is sometimes two-celled ; and in 1895 Mottier 30 stated 

 that the archesporium of Delphinium tricorne is sometimes 

 more than one-celled, that of Ranunculus abortivus one or two- 

 celled, that of Caltlia palustris two to five-celled, and that of 

 Anemonella thalictroides probably many-celled. Later Coul- 

 ter 38 found the archesporial cells of several species of Ranun- 

 culus varying in number from one to thirteen (Fig. 25), and 

 the several-celled archesporium of Helleborus cupreus is fa- 

 miliar. It is evident, therefore, that the Ranunculaceae, while 

 ordinarily producing a one-celled archesporium, show a strong 

 tendency to an increase in the number of cells. 



These three groups, Amentiferae, Ranunculaceae, and Rosa- 

 ceae, are recognized as among the more primitive members 

 of the Archichlamydeae, and the temptation is strong to con- 

 clude that the many-celled archesporium is a primitive feature 

 of the Dicotyledons. This may be true in a very general 

 sense, for no large groups have shown such a general tendency, 

 but account must be taken of the same phenomenon in other 



* Used in this connection only as a term of convenience to include several 

 of the more primitive orders of Archichlamydeae. 



