264 



THE AMERICAN NATURALIST. [Vol. XXXVIII. 



1 1, 12). Those spring tracheids which he in direct contact with 

 the summer wood of the previous year, often exhibit this feature 

 with great prominence, but it may also extend radially through 

 several successive tracheids. This is undoubtedly a primitive 

 character, and in the one case cited it possesses some value for 

 the purpose of specific differentiation, but in general terms, the 

 occurrence of bordered pits in such positions is of so sporadic a 

 nature as to give this feature no well defined value, either for 

 taxonomic or phylogenetic purposes. It may, nevertheless, be 

 stated with respect to the pits on the tangential walls of the 

 tracheids in general, that in their distribution they distinctly 

 conform to the law governing similar structures on the radial 



Reference to Cordaites acadianum shows that in the multise- 

 riate pits of the hexagonal form, these structures always preserve 

 the spiral arrangement characteristic of the structures from 

 which they were derived (Fig. 4), and this conformity also 

 extends to the direction of the spirals which generally ascend from 

 left to right. The general law in this respect has already been 

 formulated so fully by De Bary (9, p. 163), as to make it 

 unnecessary at this time to enter upon its consideration more in 

 detail, beyond a reference to one or two special features and 

 some apparently exceptional cases. While the spiral arrange- 

 ment is always typical in such genera as Cordaites, Agathis, 

 Araucaria, etc., it is not obvious in those cases where the pits are 

 strictly uniseriate and often remote from one another. Nor is 

 it apparent at first sight in those cases of 2-seriate pits where, 

 as in Cupressoxylon dawsoni from the Cretaceous, Larix amcri- 

 cajia, Sequoia and various species of Pinus, the pits are always 

 paired off in such a way that the axis of each pair is at right 

 angles to the axis of the cell (Fig. 7). Two explanations are here 

 possible: (i) the spirals are in reality 2-seriate, and they are 

 projected through the alternate members of the two rows of pits, 

 or (2) the disposition of the pits represents an extreme phase in 

 the flattening of the original spirals conformably to a higher 

 type of development. This latter view, which seems the more 

 reasonable, is in direct harmony with De Bary's law, while it 

 receives additional support from the form and direction of the pit 

 orifice. 



