22 Floristik, Geographie, Systematik etc. 



Colobanthus monticola, Petrie. Aciphylla oreophila, Petrie, Co- 

 prosma Astonix, Petrie, Celmisia Cockayniana, Petrie, C. Broweana, 

 Petrie, Gentiana Matthewsii, Petrie, Euphrasia Laingii, Petrie, E. 

 Townsoni, Petrie, Pimelea Croshy-Smithiana, Petrie, Festuca tnulti- 

 nodis, Petrie and Hackel. M. L, Green. 



Poppelwell, D. L., Notes on the Plant CoveringofCod- 

 fish Island and the Rugged Islands. (Trans. N. Zealand 

 Inst. XLIV. p. 76-85. 1 pl. 1912.) 

 These Islands lie off Stewart Island N. Z. and are the first 

 barrier met by south-west storms. The notes are the result of a 

 Short Visit. A list of 111 species collected is given, and the chief 

 plant associations are briefly outlined. Forest Covers most of Cod- 

 fish Island, with Dacrydium cupressitium, Metrosideros liicida, and 

 Weinmannia racemosa as conspicuous trees, beneath which ferns 

 form the principal undergrowth. The coastal scrub of the island is 

 mainly an association of Senecio rotundifolius, Olearia Colensoi, and 

 O. ayigustifolia. Sand dunes occupy portions of the shore, but rocky 

 cliffs predominate. The author's conclusion is that the flora of the 

 island does not differ materially from that of the mainland a mile 

 away, except that the number of species is limited. Wind is the 

 principal factor in determining the distribution: thus in the coastal 

 scrub 0. angustifolia replaces the other shrubs in the more exposed 

 places, and this species is in turn displaced leaving almost bare 

 rock. The 3 figures are photographs of the district. W. G. Smith. 



Raunkiaer, C, Measuring apparatus for Statistical in- 

 vestigations of plan tfor mations. (Bot. Tidskr. XXXIII. p. 

 45—48. 1 textfig. Köbenhavn 1912.) 



As previously pointed out (see Bot. Centralbl. 113. p. 662) Raun- 

 kiaer determines the frequency of the species in the formations by 

 means of a certain number (as a rule 50) of samples taken at 

 random; he originally employed a square frame including an area 

 of yV Square metre, and this frame was thrown at random in the 

 formation which was to be investigated. In the samples of the for- 

 mation thus obtained all species were noted down, and the degree 

 of frequency of a species is expressed by the number of samples 

 in which in was found. 



In the present paper another apparatus for the limitation of the 

 samples is described. It consists of a ring which b}'' means of a 

 screw can be fixed on a Walking stick; a piece of metal is fixed into 

 one side of the ring, it has a screw-thread into which a metal rod 

 can be screwed, and the length of this rod is equal to the radius 

 of a circle including -^-^ square metre. 



The mode of proceeding is this: The stick is with stated inter- 

 mediate Spaces stuck into the ground, and every species occurring 

 within the circle described by the tip of the rod is marked in the 

 record-table. Of course rods of different lenghts can be employed. 



This apparatus has been found to be more practical in use than 

 the frame. Ove Paulsen. 



Rendle, A. B., Notes on Tropical African Convolviilaceae. 

 (Journ. Bot. L. p. 253—254. Aug. 1912.) 

 2 new species are described: Ipomoea alpina and /. Kassneri. 



M. L. Green. 



