290 



SKETCH OF THE CONIFERS OF JAPAN. 



above, and fig, 49 of an opened cone) ripening in the second 

 year, growing on a very short stalk, ligneous, subhorizontal or 

 dependent, smaller than those of P. sylveMris, rounded at the base, 

 from that upwards conical, obtuse. Scales ligneous cinereous 

 brown, linear-oblong, thickened at the apex, rhomboidly truncate 



« 



Fig. 50. 



\ ¥ 



Fig. 49. 



Fig, 51 



areolate, mth the transrerse umbo near tlie upper margin (fig. 50). 

 Seeds, two attached to each scale, elliptic, with the apex obliciuelj 

 truDcate, with an obtuse pale cultriform wing slightly shaded 

 with a darker hue (fig. 51). Test or nut crustaceoiis and pale 

 brown, without spots. 



As already mentioned, this tree is so closely allied to P. If^is- 

 soniana that it is not without great hesitation it has been here 

 recorded as distitict. The only characters which can be cited as 

 distinguishing them are slight, and unless these are singularly 

 constant they appear hardly of sufficient value to be reckoned 

 as specific. It is true that if there be no transitional passages 

 between the type specimens of each brought home by Mr, 

 Fortune and Mr. Veitch, there is no difficulty in at once 

 identifying them. The foliage of P. Massoniana is longer, 

 stouter, and more rigid than that of P. dcnsiflora ; there are 

 one or two more rows of storaata upon the leaf of P. Massoniana'; 

 but this, in a number like 18, is no great matter. The cones of 

 both are nearly of the same size, but those of P. Massoniana are 

 slightly larger and comparatively longer and more acute, and 

 someu-Jiat tncurtcd, which is not the case with P. lUnsiflora. 



