402 Proceeding's of the Br'ttish Association. 



branches of the natural family of Coniferae. He stated that in 

 these branches the pith is always nearer the upper than the 

 under side ; that the upper side is of a paler colour than the 

 under side ; that the upper side is softer and less dense than the 

 under side ; and that, whilst the upper side has a structure 

 similar to that of the stems, the under side has a structure so 

 different in all the three principal sections, that without ocular 

 proof, no one could imagine it to belong to the very same 

 branch. The transverse section has the partitions forming the 

 net-work of the under side considerably thicker than those of the 

 upper side. The vessels or openings of the former are conse- 

 quently smaller than those of the latter, and hence the greater 

 solidity of the wood on the under side. The longitudinal sec- 

 tion, parallel to the radial partitions of the under side, have 

 smaller, less numerous, and more obscure discs than those in the 

 upper side. The vessels or spaces containing the discs in the 

 under side present numerous decussating fibres, which do not 

 occur in the upper side ; and these fibres also occur in the lon- 

 gitudinal concentric section of the under side. The difference 

 of the structure in the upper and under sides of these branches 

 may be seen very accurately delineated in the third plate of the 

 sixteenth volume of this Journal. The three upper figures are 

 representations of the structure of the upper side, and the three 

 under figures are representations of that of the under side. The 

 branches of ten different species of pines were examined, and 

 the same structure was observed in them all, although in some 

 it is better defined than in others. The peculiarity of the struc- 

 ture in the branches was stated as an additional proof of the ab- 

 surdity of attempting to constitute new fossil genera, on the 

 supposition that a single slice of a Scotch or spruce fir is charac- 

 teristic of the whole family of conirera\ In conclusion, Mr 

 Nicol observed, that an accurate knowledge of the anatomical 

 structure, \n^uld soni^limcs enable the botanist to classify aright 

 when the external characters might leave him in di)ubt, and in 

 proof of this was observed, that, had the structure been known, 

 the miro of New Zeali.nd would not have been represented as 

 a podocarpus. The structure of that tree bears no resemblance 

 to that of any of the coni^era^, its character being that of a true 

 dicotyledon. It was also mentioned that the Tasmannia dipeta- 



