48 Bibliographical Notices. 



We must now enter a little into detail. The author has paid pe- 

 culiar attention to the elucidation of the species included in the dif- 

 ficult genera Bromus and Poa. In the former he has successfully 

 applied some new characters to the distinction of those difficult 

 species, B. mollis, B. racemosus, B. arvensis (or rather commutatus), 

 and B. seculinv.s ; and in the latter he has used throughout the genus ' 



characters, founded upon the webbed flowers, the nerves of the pa- \ 



lese, the sheaths of the leaves, the ligules, and the joints of the stem, f 



which had not previously been employed in more than a few species. i 



The result has been that several forms of Poa which appear to have | 



great claims to specific rank have been detected, and it seems pro- 

 bable that when all the British plants belonging to this genus have 

 been examined with equal care, that several other additions will re- 

 quire to be made to our list. Dr. Parnell distinguishes P. ceesia, P. 

 inontana (a new species), and P. Balfouri* (n.s.) from P. nemorulis, 

 and P. polynoda (n.s.) from P. compressa. He considers his P. Mon- 

 tana to be the P. nemoralis, var. montana, of Koch's ' Synopsis '; but 

 for this there do not seem to be sufficient grounds, and the Scottish 

 plant is probably a species not before recorded. 



Three genera not usually recognized in this country have been 

 adopted, namely, Amenagi'ostis, Bucetum, and Trisetum. I'o the for- 

 mer, which includes the Agrostis spica-venti alone, we have nothing 

 to object except its name, which ought to have been Apera. Con- 

 cerning the other two it will be necessarj^to speak at greater length. 

 To Bucetum (a name invented by Dr. Parnell) he refers the Festuca 

 elatior, F. pratensis, F. loUacea, and F. gigantea of authors. All 

 these we refer to Festuca, considering the three former as belonging 

 to one variable species. The only distinctions that we can detect be- 

 tween Bucetum and Festuca are, that the awned midrib is not at- 

 tached to the palea quite to its summit in the former, and that the 

 radical leaves are broader than those of the stem ; whilst in Festuca 

 the awn is (usually) quite terminal, and the stem leaves are broader 

 than the radical. We do not consider this as a sufficient reason for 

 constituting a new genus. 



In Trisetum, the third genus to which we have referred, our au- 

 thor places Avena pratensis (including as varieties A. alpina and A. 

 planicuhnis) and A. pubescens in addition to A. flavescens, which has 

 been often referred to that genus. As he has not contrasted the 

 characters of Avena and Trisetum, it is rather difficult to ascertain 

 upon what he would found their distinctions ; indeed the short ge- 

 neric definitions do not afford any tangible point, except that Avena 

 is included in the section "calyx containing two florets," and Trise- 

 tum in that with " the calyx containing three or more florets." Now 

 this would exclude from the genus Avena several true Oats, such as 

 A. sterilis, which often has four florets ; A.fatua, in which three florets 

 is as common as two; and A. nuda is often, if not usually, three-flow- 

 ered. A character may certainly be found in the fewer ribs of the 

 glumes and palese of Parnell's Trisetum, but that cannot be a suffi- 



* This new grass was figured and described in vol. x. p. 121 of this Jouriial. 



