PINACEAE. — JXTNIPERU3 59 



wise identical is fairly good proof that this plant is an alpine form of the Chinese tree. 

 Juniperus recurva Hamilton differs in its subulate, closely appressed leaves, recurved 

 brunches which singularly resemble those of Lycopodium, and in its much larger 

 (1 cm. long) fruit. That a dwarf form of Hamilton's plant also occurs is shewn by 

 several specimens before us of which we may cite one collected by G. A. Gammie 

 m the Luchung valley, alt. 4000 m. August 1892, and described as a small shrub. 

 This specimen like the others we have seen is typical J. recurva Hamilton, except 

 that the branches are shorter, the leaves much reduced in size and the inteinodes 

 much shorter — differences naturally to be expected in an alpine form. Probably 

 this shrub is common at high altitudes in Silckim and elsewhere, and may some- 

 where grow in association with /. squamata Lambert. This would account for the 

 confusion which has existed in the identity of the two species. Sikkim marks the 

 western limit of distribution of several Chinese plants, and the neighbourhood of 

 Tachien-lu the eastern limit of certain Himalayan plants. 



Juniperus densa Gordon, is usually quoted as a synonym of /. squamata Lam- 

 bert, but Gordon (PiJielum, Suppl. 32, [1862]) describes his plant as having a 3- 

 seeded fruit and therefore it must be referred to some other species, probably to J 

 communis Linnaeus. There is however in gardens a Juniper known as J, recurva, 

 var. densa, which is the true J. squamata Lambert and, further, a specimen before 

 us from Gordon's herbarium and labelled /. densa, is unquestionably a form of 

 •/. recurva Hamilton. Apparently two or three Jumpers are in cultivation under 

 Gordon's name /. densa. 



From the description we strongly suspect that J. morrisonicola Hayata (in 

 Gard. Chron. ser. 3, XLIII. 194 [1908]) is really J. squamata Lambert. Specimens 

 from Mt. Omei collected by Faber (Masters in Jour. Linn. Soc. XXVL 543 note 

 [1902]) in all probability belong here also, as we have seen the plant growing abun- 

 dantly on the summit of that mountain. 



Pictures of this shrub will be found under Nos. 226, 228 of the collection of 

 Wilson's photographs and in his Vegetation of Western China, Nos. 264, 265. 



Juniperus squamata, var. Fargesii Rehder & Wilson, n. var. 



Juniperus Fargesii Komarov in Herb. Paris. 



jrnmi 



saepe plus minusve patula et minus congesta quam in typo, lanceo- 

 lata, 4-12 mm., plerumque 7-9 mm. longa, circiter 1 mm. lata, acumi- 

 iiata, spinulosa, pallide v. coenileo-viridia, supra saepe glauca. Fructus 



ovoideus, anice ohtnsns. .^Wi tYim. Tnnfms. obscure niirro-coeruleus. 



"'--.^cuK, apice ODtusus, b-K> mm. longus, oDSCure mgro-cueruieua. 



Western Szech'uan: west and near Wgn-ch'uan Hsien, wood- 

 lands, alt. 2000-2500 m., October 1908 (No. 1021); west of Kuan 

 Hsien, woodlands, alt. 2000 m., June 21. 1908 (No. i02i»); Wa-shan, 



v^liengkou Ting, P. Farges (No. 153, type of J. Fargesii Komarov). 



, "^^is is the arborescent form of the type and is not uncommon in woodlands 

 throughout the Chino-Thibetan borderland. Usually it is a small tree from 5 to 

 ^-5 m. tall, but occasionally larger trees are met with. We saw and photographed 

 one magnificent specimen 28 m. tall, with a trunk girth of 7 m., but we never met 



