652 SKETCH OF THE CONIFEKS OF JAPAN. 



bark resembles that of the Cedar, On the other hand, the large 

 broad flat leaves are unlike anything previously known among the 

 Larches. The cones and their scales hear no resemblance to 

 those either of the Larch or Cedar, nor for that matter to the 

 cones of any other known tree. Its seeds and wmgs have no 

 relation to these parts in the Larch ; they rather resemble the 

 seeds of a Picea, to which the seeds of the Cedar are not without 

 alliance, and like them are largely filled with turpentine. If we 

 attempt to ascertain its place by a reference to the present syste 

 matic works on the Coniferm, we find that no section will receive 

 it. For instance, take Endlicher, perhaps the best systematist on 

 the Coniferm, It will not rank with Txmja, AbieSy Larix, nor 

 Cedrus, because the scales of its cone are not persistent, and it will 

 not rank with Picea, because the leaves are not solitary nor persis 

 tent but in verticillate deciduous clusters. Although, as in Picea^ 

 its scales are not peristent, this latter particular is of a totally dif- 

 ferent nature in the two. In the Pice^s there is a round core like 

 a branch, to which the scales are attached by a very small narrow 

 pedicle, which when dry becomes brittle, and the scale drops off 

 from its own weight. To keep the scales together, therefore, the 

 cone grows erect and rests on the horizontal body of the branch. 

 In Ps, Kampferi the core of the cone is quite different; the axis is 

 merely a slender accumulation of the single fibres, which proceed 

 to the base of each scale surrounded by some corticaceous matter. 

 It also grows pendent, a position totally repugnant to the Picem. 

 On the whole there is no doubt that it will enter into none of the 

 sections or sub-genera yet established. The pulvini, to which we 

 attach great importance as a sectional character^ indicate its place 

 to be next the Larch. It may, indeed, be treated as a section 

 of that sub-genus, but on the whole we agree with Mr. Gordon 

 that it deserves to be erected into a sub-genus, although we dis- 

 like the name under which he has attempted to do so. 



After what has been above said under A. Leptolepis, it is scarcely 



necessary to correct Mr. Gordon's statements regarding these 

 trees. 



