Primulacee.] . THE HIMALAYAN MOUNTAINS. 309 
wards to the Mediterranean, the Caucasus, and even the Himalayas, where only a 
single species (P. ce@ruleum according to Dr. Wallich) is found in Kemaon and at 
Kotgurh ; but the specimens which I have from the Choor Mountain and the Borang 
Pass, as well as from Cashmere, agree better, according to Mr. Don, with P. gracile 
of Fischer, from Siberia. In addition to this, there is another genus found on lofty 
situations in the Himalaya, which is referable to this family, on the authority of 
Mr. Bentham, who has named and described so many of the new genera and species 
of Polemoniacee, and has with his accustomed kindness favoured me with the following 
character of those of Cyananthus, which with its 5-celled ovary and 5-lobed stigma, 
agreeing with some Campanulacee, differs from them in its inadherent calyx. 
CyaNnantuus, WALL. 
“* Calyx inferus tubuloso-campanulatus semi-5-fiduslobis non membranaceis. Corolla subinfundibu- 
liformis tubo amplo limbo 5-fido. Stamina 5 ad basin corolla inserta ejusque lobis alternantia. Anthera 
ovario adpressa connatave. Ovarium liberum 5-loculare. Ovula in quoque loculo plurima. Stylus simplex. 
Stigma 5-lobum lobis ovatis patentibus. Capsula oblongo-conica valvis 5 loculicide-dehiscens.—Herbe 
annuz parum ramos procumbentes vel adscendentes. Folia alterna simplicia. Flores pauci solitavii 
sepiusterminales. Calyces nigro-villosi. Corolla speciose caerulea. Semina oblongo-linearia. 
1. C. integra, (Wall. Cat. Herb. ind. n. 1472), foliis ovato-oblongis integerrimis serrulatisve, 
Hab. Kemaon. Waillich. Kedarkanta. Royle. 
2. C. lobata (Wall. 1. c. n. 14°78), foliis cuncatis inciso-lobatis, 
Hab. Kemaon et Gossainthan, Wallich. Kedarkanta. Royle. 
These plants differ from the other genera of the order in their 5-celled ovarium and 5-lobed stigma, 
but in every other respect they appear more nearly connected with Polemoniacee than with any other 
tribe.” 
131. PRIMULACE. 
The Primulacee, well known from affording: so many ornamental and favourite flowers, 
are most closely allied in structure to Myrsinee, from which, however, they differ 
much in habit. Dr. Lindley has remarked, that ‘‘ they are nearly allied to all the 
regular monopetalous orders, with capsular fruit, especially Solanee and Gentianee.” 
They are usually placed next Lentibularee, which are more closely related to Scrophu- 
larine@. Hence it is difficult to place this family appropriately in a linear series; but 
if we suppose the series as returning into itself, and forming a plexus of circles and 
natural classes, the Primulacee will serve to connect several together. 
Ornamental as are all the known Primulacee, they are no where more so than on the 
coldest and most exposed of the Himalayan peaks, and at a season when returning 
warmth has but just begun to displace the scene of wintery desolation. The plants of 
this family are chiefly diffused through the temperate and cold parts of the northern 
hemisphere, both in the Old and New World, where they are found in moist, shady, 
and frequently mountainous situations. But they exist also in the southern hemisphere, 
as Primula in the Straits of Magalhaens, and I believe also on the mountains of Van 
Diémen’s Land ; Lysimachia and Centunculus in New Holland, and Anagallis in Chili 
and Peru. 
In India, the only plant of this family which is generally diffused through the plains, 
but this only during the cold weather, is Anagallis arvensis, found also in the Hima- 
layas, 
