60 REVISION OF PRIMULA. 



general cultivator. By means of keys, constructed on a new and 

 very ingenious system, the position of a species or a genus in the 

 natural order heing seen at a glance. The European hybrids will 

 he found rather complicated by the uninitiated, owing to so much 

 intercrossing said to having taken place between the species, which 

 can only be decided after long and careful study of the living plants. 



Dr. Pax differs most from the English botanist in the definition 

 of species ; his views of the officinalis section, however, nearly 

 coincide with ours, and we are glad to see that he does not believe 

 in the endless hybrids said to be derived from these common 

 British plants, elatior, officinalis, and vulgaris, the greater number of 

 which he properly places as synonyms or mere varieties. The 

 results of H. C. Watson and Hofmann in crossing these plants are 

 very interesting : the former, it is stated, raised elatior, acaulis, and 

 officinalis fi-om the same seed, and Hofmann is said to have changed 

 P. elatior into P. officinalis in six generations ; but here Dr. Pax 

 cautiously observes that these experiments might not prove good in 

 the face of a strict criticism. 



The farinosa section is defined on the old plan, with the exception 

 of a few regarded in this country as varieties being claimed as 

 species, and a few, such as P. lepida Duby, being placed as varieties 

 of P. farinosa. The greatest change, however, is in the nivales. 

 Dr. Pax observes that amongst the Asiatic species are four types, 

 one Siberian, one Himalayan, and two limited to Sikkim. The 

 Siberian type, P. nivalis Pall., with crenate, serrate, rarely entire 

 leaves, has a number of forms — one in the Caucasus, P. Bayneri; 

 one in Turkestan (var. farinosa) ; ^loorcroftiana, in E. Siberia, 

 which there attains its largest development ; the variety pimiila 

 Ledebour, which has been raised to specific rank ; and finally, 

 P. purpurea Eoyle. P. Stuartii is retained as a distinct species, and 

 Moorcroftiana, x>^^n^urea, lineariloha, and macrocarpa are transferred 

 from varieties of this species to P. nivalis, which seems a very 

 intelligible conclusion. The chief diflerence between P. nivalis and 

 P. Stuartii is in the colour of the flowers ; all these varieties have 

 purple flowers, and, although they form connecting links between 

 the two species, the difficulty seems to have been surmounted on 

 the most reasonable grounds. 



The European species have been well done, but with such a 

 mixture of hybrids as to confuse the minds of all ordinary mortals 

 tryiug to unravel them. The practice by nurserymen on the 

 Continent seems to have been for a long time past to assign the 

 parentage to a plant on msufficient grounds, without in the first 

 instance testing its truth. The plant known in English gardens as 

 P. nivea here has been defined ; Dr. Pax places it under hirsuta, 

 but with this we cannot agree. P. nivea as grown in our gardens 

 is more robust than hirsuta, and the scapes and calyx, and some- 

 times the leaves, are always covered with farina, which points to 

 P. pubescens ; and this is certainly where it belongs, as anyone can test 

 for himself by raising pubescens as cultivated in gardens from seed. 



On the whole we have much pleasure in adding our testimony 

 to the value of Dr. Pax's monograph ; it has been a labour of love, 



