72 A. H. Graves, 



fig. 1, a and b), which continues the growth of the stem. Of these 

 branches, that in the axil of the upper leaf (Text-fig. 1, a) is invariably 

 of stronger growth, exceeding considerably the branch on the other 

 side of the floral axis. These axillary branches, after forming a few 

 leaves, terminate in flowers again, and the process of branch form- 

 ation is repeated. In this way, since the main axis is continued 

 by the stronger branch, the system in the region of the flowers is 

 sympodial. In most instances, however, the weaker axillary branch 

 has a fairly vigorous development, as in Text-fig. 1, in which case 

 the branching approximates a false dichotomy, and a characteristic 

 fan-like form is thus often produced. 



In Text-fig. 1 the branch a develops two ordinary leaves before 

 it terminates in a flower. This brings the upper of the pair of sub- 

 floral leaves at the right side instead of the left, as at I. Such 

 an arrangement as this is not as common as is the continual pro- 

 duction of the stronger branches on the same side of the stem, 

 but in this way also the fan-like form may be attained. 



2. Comparative Study of the Branching of the Potamogetonaceae. 



In Zannichellia, Campbell (1897, pp. 38 fl.) finds that the apex of 

 the stem divides into two equal parts, one of which develops into 

 the female inflorescence, while the other continues as the main 

 axis of the stem. Farther on he says " the inflorescence is the 

 result of the dichotomy of the main shoot, whose other member 

 continues the growth of the axis." So that, although a sympodium 

 is the ultimate result, it is attained by a dichotomy of the growing 

 point, one of the branches becoming a male or female flower. 

 Schumann's (1892) interpretation had been somewhat different from 

 this, maintaining that the female inflorescence terminated the main 

 axis and that the growth of the shoot was continued by an axillary 

 branch arising in the upper of the two subtending leaves. The 

 position of these leaves much resembles that of the subfloral leaves 

 of Ruppia, so that if Schumann's idea were correct, the inflorescent 

 branch system of Zannichellia would be a sympodium similar in 

 origin and development to that of Ruppia. 



Campbell's figures, however, show a vigorous, simultaneous de- 

 velopment of the primordia of both flower and main axis, which 

 lend to his theory a great deal of weight. In Ruppia the primordium 

 of the axillary shoot arises much later than that of the flower, so 

 that such a dichotomy as Campbell claims is here impossible. 

 (See PL IX, figs. 54-56.) 



