Petrie. — On Poa breviglumis, Hook. f. 197 



Art. XIX. — On Poa breviglumis, Hook. f. 

 By D. Petrie, M.A. 



[^Read before the Auckland Institute, '22nd November, 1909.] 



The writer has long entertained a suspicion that the native grass Poa 

 breviglumis. Hook, f., was not really distinct from what is taken to be the 

 Poa imbecilla of Forster. Hooker's species was originally published in the 

 " Flora Antarctica," vol. i, pp. 101 and 102 (1844), and was founded on 

 specimens collected on Campbell Island. The name Poa imbecilla was 

 applied by Forster in 1786 to a New Zealand grass, but he pubhshed no 

 diagnosis or description of it. A number of years later Forster's name was 

 assigned by Allan Cunningham, in his " Prodromus " (1836), to the Poa 

 now so named. 



In the " Flora Novse-Zelandioe " (1853) Hooker includes and describes 

 Forster's Poa imbecilla, accepting Cunningham's identification of the plant, 

 but he does not in any way refer to its likeness to his own Poa breviglumis. 

 This may have been due to the fact that the forms of Poa imbecilla, Forst., 

 that occur in the more northerly parts of New Zealand are not typical of 

 the prevalent state of the species. In the " Handbook of the Flora of 

 New Zealand " (1867) both the above Poas are included, and the author 

 concludes that both grow on the main islands, from which he had at first 

 supposed Poa breviglumis to be absent. Though it is unlikely that he had 

 good or abundant material to work on when the Handbook was in prepara- 

 tion, he recognised the close affinity of the two species, and even suggested 

 that his Poa breviglumis might be a variety of Poa imbecilla, Forst. He 

 further mentions that Banks and Solander referred the mainland forms of 

 his Poa breviglumis to Poa imbecilla. 



In his excellent " Manual of the New Zealand Flora," published in 1906, 

 Mr. Cheeseman maintains the two species as distinct, but refers all the forms 

 found on the main islands to Poa imbecilla, Forst. He mentions also that 

 he had seen only a fragment of one of Hooker's specimens from Campbell 

 Island, and some two or three others collected by Kirk and Chapman on the 

 Auckland Islands. All these specimens I have, through Mr. Cheeseman's 

 kindness, been able to examine. 



The Philosophical Institute of Canterbury's expedition (November, 1907) 

 for the exploration of the subantarctic islands of New Zealand secured no 

 specimen of the grass, the visit being undertaken at too early a season. 



In January of this year Mr. B. C. Aston, a most enthusiastic and capable 

 collector, visited the subantarctic islands, and made a fine collection of all 

 the grasses he met with, including a number of specimens of Poa brevi- 

 glumis, from Enderby Island (in the Auckland Group) and from Campbell 

 Island. These have been entrusted to me for examination, and the speci- 

 mens appear to make the position of the plant quite clear. From a study 

 of this material I conclude that Poa breviglumis is merely a form of Poa 

 imbecilla, and that it is doubtful if it should be ranked as a distinct 

 variety. 



The descriptions in the " Flora Antarctica " and in Cheeseman's Manual 

 lay emphasis on the prominent nervation of the flowering-glumes. I am 

 unable to find in these glumes of Poa breviglumis anything to distinguish 



