THUIDIUM RECOGNITUM AND ITS ALLIES 191 



rocks referred by me to that species, in the article previously 

 mentioned, being the plant described by Limpricht as T. pseudo- 

 tamarisci, but reduced by Eyan and Hagen — certainly correctly — 

 to a var. of T. Philiberti, of which it is merely a robust, usually 

 tripinnate form. 



One or two points will make the matter clearer and at the 

 same time will explain in part how the difference of view has 

 arisen. The Continental view of the species is undoubtedly the 

 correct one, but the distinctive characters attributed to them have 

 been, I think, in some respects erroneous, and this has led to 

 misunderstanding. 



In the first place, T. Philiberti is by no means confined to wet 

 places, as stated by Dr. Best, but is quite as frequently found on 

 dry banks. In this case the hyaline piriform points of the stem- 

 leaves are but little developed, and are easily lost, so that their 

 presence is often scarcely to be detected. 



In the second place, T. recognitum has not plane-margined 

 stem-leaves. The margins are indeed much less strongly recurved 

 than in the allied species, and very frequently only one margin is 

 reflexed, and that slightly ; but in all the plants received from 

 Continental authorities as the true T. recognitum, I have never 

 found the margins actually plane, except, possibly, in a single leaf 

 here and there. 



Nor does the nerve actually fill the acumen, as is frequently 

 stated. It is quite true that it remains broad and stout, and 

 reaches very nearly if not quite to the apex, and is very different 

 in appearance from the structure in T. Philiberti and T. delica- 

 tulum, but this is due not only to the greater extension of the 

 nerve, but to the fact that the cells of the lamina in the acumen 

 become elongate, narrow and incrassate, resembling the elongate 

 cells of the nerve itself, and thereby rendering the delimitation of 

 the latter less easily observed. In the two allied species the cells 

 of the acumen, while often slightly elongate are not markedly so, 

 and are scarcely altered in character from those of the rest of the 

 lamina, so that the acumen is of a quite different texture from 

 that of T. recognitum. 



The size of the stem-leaves, frequently given as a specific test, 

 does not appear to me reliable ; thus I frequently find T. delica- 

 tulum with stem-leaves quite as small as in T. recognitum, 



Another character relied on by some authors is the abrupt 

 narrowing of the stem-leaves into the acumen supposed to be 

 characteristic of T. Philiberti, as compared with a gradual nar- 

 rowing in T. delicatulum, but I have found this, in practice, quite 

 unreliable. 



For all that the two plants are really distinct, and are recog- 

 nized when once known without great difficulty. The distinguish- 

 ing characters may be pointed out thus : — 



T. recognitum is usually a plant of calcareous woods, and is 

 frequently fertile — much more so, probably, in relation to its 

 rarity, than T. tamariscinum. The stem-leaves are small as com- 

 pared with well-developed T. Philiberti and — usually — T. delicti- 



