46 Transactions. 



His opinion was that the two plants, while undoubtedly within the circum- 

 scription of Parsonsia capsularis R. Br., differ in certain respects, and that 

 my northern plant should be accorded a varietal designation. He further 

 suggested that the small-flowered form should also be named as a variety, 

 the conception Parsonsia capsularis being evidently an aggregate of three, 

 and probably more, distinct forms. 



I take this opportunity of heartily thanking Dr. L. Cockayne, F.R.S., 

 for the assistance he has afforded me in this connection. 



Parsonsia capsularis R. Br. var. parviflora Carse var. nov. 



Frutex tenuis ramosus scandens. Folia valde polymorpha, plantarum 

 juvenilium 30-75 mm. longa, angusto-linearia, lanceolata, vel spathulata, 

 integra, sinuato- vel inaequi-lobata ; plantarum maturarum variabilia, 

 angusto-linearia, 25-100 mm. longa, 2-5 mm. lata, oblonga vel oblongo- 

 lanceolata, 25-75 mm. longa, 12-5-20 mm. lata, obtusa vel subacuta, 

 coriacea, margine plerumque integerrima. Cymi pauci- vel multi-flori, 

 axillari vel terminales. Flores parvi + 3 mm. longi. Lobi calycis corollae 

 tubum aequantes. Corolla campanulata, lobi revoluti. Antherae exsertae. 



This is the slender climber, with small creamy flowers, which is not 

 uncommon in the localities stated above. In the North Island, so far as 

 I am acquainted with it, it usually occurs in open hilly woods. I take 

 it to be the original Periploca capsularis Forst. f., but before this can be 

 finally determined specimens must be examined from the neighbourhood of 

 Queen Charlotte Sound. 



Parsonsia capsularis R. Br. var. grandiflora Carse var. nov. 



Frutex ramosus scandens, quam var. parviflora robustior. Folia 

 similiter polymorpha. Cymi plerumque multiflori. Flores majores, rubri, 

 vel luteo-rubri, nunquam rosei, 4-6 mm. longi. Corollae tubi lobos calycis 

 equantes vel superantes. 



North Island. Great Barrier Island : Colonel Boscawen, per Mr. Cheese- 

 man ! Whangarei district, Mangonui County : H. C. Usually in damp 

 lowland situations. 



A much more robust liane than var. parviflora. In most of my speci- 

 mens the leaves are more or less ovate-lanceolate, but heterophyllous 

 forms are not uncommon. At first glance this plant may easily be mis- 

 taken for Parsonsia heterophylla A. Cunn., but the exserted anthers at 

 once place it under P. capsularis. The colour of the flowers varies from 

 creamy-yellow to a red like tiie flesh of a pumpkin. Mr. Cheeseman 

 describes Colonel Boscawen's specimen as having " orange-red flowers." 

 It is undoubtedly the same plant as occurs from Whangarei northward. 

 From var. rosea it differs in the more robust habit and larger differently 

 coloured flowers. 



It is quite possible that the above variety may be Colenso's P. ochracea, 

 though from his description (Trans. N.Z. Inst., vol. 22, 1890, pp. 480-81) 

 I should say his was the small-flowered plant. His herbarium specimens 

 appear to have been mislaid, so that at present it is not possible to compare 

 them. 



