T. Kirk. — A Revision of the Neiv Zealand Gentians. 333 



ful in elevated regions of both temperate and tropical zones, 

 even reaching the snow-line, upwards of 16,000ft., they become 

 rare or absent in higher latitudes. Generally the New Zea- 

 land forms are restricted to alpine or subalpine situations, but 

 G. saxosa, G. cerina, G. antipoda, and G. antarctica descend 

 to the sea-level, and are never found out of the reach of sea- 

 spray, the first named being restricted to Foveaux Strait, the 

 others to the Antarctic Islands. G. montana, G. bellidifolia, 

 and perhaps G. pleurocjynoidcs are found in the North 

 Island, the first extending from the East Cape southwards 

 to Stewart Island, the second only to Otago, the third to 

 the Chatham Islands and Otago ; and the remaining three 

 species are restricted to the South Island, two of them 

 being remarkably local. All the species appear to be endemic 

 except G. montana, G. 'p^^'ogynoidcs , and G. bellidifolia, 

 which are also found in Tasmania and the southern part of 

 Australia. 



Bearing in mind the extreme rarity of white-flowered 

 gentians in other countries, it is remarkable that all the local 

 species primarily produce white flowers, although G. antipoda, 

 G. cerina, and G. antarctica var. imbricata exhibit also various 

 shades of red, purple, and violet, or occasionally white wuth 

 vertical streaks of red : the other species are pure-white, or 

 rarely exhibit a pale-lemon or sulphur tinge, which, however, 

 never merges into a true vellow, such as that of Gentiana 

 lutea, or the common buttercup. Sometimes, however, this 

 very pale tint becomes greatly intensified in herbarium speci- 

 mens, and has led to the erroneous idea tbat some of the New 

 Zealand gentians have yellow flowers." Dr. Jameson, in his 

 "Botanical Notes on the Flora of the Andes of Peru and 

 Columbia,"! remarks, " Of sixteen species of gentian with 

 which I am acquainted, one-half are red, four purple, two blue, 

 one yellow, and one white," thus offering a strong contrast to 

 the New Zealand species, all of which are primarily white. 



In the following descriptions an attempt has been made to 

 differentiate tbe more striking permanent departures from the 

 typical form of the species as varieties, but the success attained 

 is less complete than could have been wished, although pos- 

 sibly sufficient has been done to prevent that sense of con- 

 fusion which often troubles the minds of young students on 

 finding a number of apparently dissimilar plants united under 

 a common name. 



* A parallel to this curious change of colour in dried gentians is 

 afforded by certain white-flowered veronicas, notably V. elliptica, V. 

 vernicosa, V. macrocayya, some forms of V. salicifolia, &c, which, after 

 being placed in the herbarium, sooner or later assume a purple tint. 



t " Flora Antarctica," i., 55. 



