39 



his views were adopted by Choisy and by Eaoul (Choix dej Plantes 

 de la Nouvelle Z^lande, p. 44). The author's observations on rather 

 limited material tend to show that not only C. tuguriorum, but also 

 C. silvatica, C. SoldaneUa, C. Dahurica and even C. spithamea are 

 referable to C. sepium. 



On the sandy beach of Gipps-Land the writer has observed the 

 transit from the trailing plant with fleshy renate leaves to the typical 

 form with hastate leaves and winding stems. New Zealand speci- 

 mina of C. SoldaneUa or reniformis, communicated by Mr. Travers 

 senior, have the ripe seeds black, smooth and about 3"' long. If in 

 reality these plants are distinct their marks of differences consist in 

 others than those hitherto pointed out. 



That C sepium, when it adapts itself to sandy saline shores, 

 should assume a prostrate growth and fleshy leaves of a somewhat 

 altered form cannot be surprising, if we remember what an effect 

 literal localities exercise on the characters of other plants. Koch 

 (Synops. Plant. Mor. German, et Helvet. ii. 568) inclines to refer C. 

 sUvatica also to C. sepium, a view fully borne out by the ex- 

 amination of Eochel's plant from the Banate in our collection. C. 

 acutifolia, sent by Prof Philippi from Valdivia, belongs also to 

 C. sepium. C. Dahurica, according to a series of specimina, named 

 by Turczaninow, flows likewise together with G. sepium and in its 

 extreme forms links, as it seems, also to that widely diffused plant 

 the North American C. spithamsea according to specimina of Dr. 

 Short's and Dr. Sartwell's collections. To a certain extent analogous 

 forms are produced by Clematis aristata in Australia. Calystegia 

 affinis from Norfolk-Island is probably referable to C. marginata. 

 The latter species (con£ J. Hook. Fl. Nov. Zeel. i 1 84, t. 48) ranges 

 from the mountain-glens in the vicinity of Port Phillip through S. E. 

 AustraUa to the southern parts of Queensland. Its capsules measure 

 hardly I" and are spherical ; the seeds are brown-black, rough and 

 about 1^"' long. This species with many other Gipps-Land plants is 

 likely yet to be found in the phytologically little explored north-east 

 part of Tasmania. That some convolvulaceous plants are subject to 

 extreme variations, generally proportionate to their adaptability to 

 varied locahties and climatic influences, is very evidently demon- 

 strated by Convolvulus erubescens. Of the changes, which the 

 external form of this plant in its almost universal range through 

 extratropical Australia undergoes, an estimation may be formed 

 when it is acknowledged, that the following plants have to be re- 

 joined with it : C. angustissimus (R. Br. Prodr. 482), C. remotus 



