408 CEDEUS. 



and the north-west Himalaya, the various parts of the three Cedars are 

 compared, the bark, wood, leaves, staminate flowers, cones, etc., and the 

 conclusion arrived at is : " that as species the three Cedars cannot be 

 distinguished, and that they must all have been derived from one 

 common stock ; * nevertheless they may be regarded as three well-marked 

 forms, which are usually very distinct but which often graduate into one 

 another." 



That they are necessarily regarded as distinct by the 

 arboriculturist and landscape gardener will be evident enough 

 from a comparison of the habit, aspect and colour of the foliage 

 of the three Cedars as seen growing in the parks and gardens 

 of Great Britain. As thus viewed the following points of difference 

 are quite obvious : 



In Cedrus Libani the primary branches are frequently long in 

 proportion to height of trunk, often of timber-like size, usually horizontal, 

 but sometimes bent downwards by the weight of their appendages, 

 regularly tabuliforrn, and the terminal growths more or less pendulous. 

 The leaves are shorter than in C. Deodara and longer than in C. 

 at/antica, and grass-green in colour. 



In Cedrus Deodara (up to forty fifty years) the primary branches 

 are usually shorter in proportion to height of trunk than in C. Libani 

 or 0. atlantica, of nearly equal length in each pseudo-\vhorl, gradually 

 shorter upwards and giving the tree a more stricirly pyramidal outline 

 than is observable in the other two ; the terminal growths are more 

 slender, more elongated with paler bark, and quite pendulous. The 

 leaves are longer and of a paler green. 



In Cedrus atlantica the primary branches are horizontal, often of 

 unequal length in the same pseudo-whorl, giving the tree a less formal 

 outline than in C. Deodara; less formally tabuliform than in C. Libani 

 and with the terminal growths mostly rigid. The leaves are shorter, 

 glaucous, sometimes of a silvery whiteness, and the cones are smaller 

 than in the other two. 



The geographical distribution of the Cedars is remarkable; they are 

 confined to three separate regions in the great mountain systems that 

 cross the eastern continent between the 28th and 38th parallels of 

 north latitude with but little interruption from the Atlantic Ocean to 

 the China Sea. The three species, here recognised as such for practical 

 convenience, occupy positions nearly equidistant, Cedrus Libani (Syria 

 and Cilicia) being in the middle with C. Deodara (Afghanistan and 

 north-west India) and C. atlantica (Algeria) east and west of it 

 respectively and separated from it by an interval of 1,200 to 1,400 

 miles, t Their habitat is thus restricted to a mountainous region with 

 a vertical range of from 4,000 to 6,000 feet elevation for the middle 



* This view of the relationship of the three Cedars was strengthened by the discovery, 

 some years after the publication of Sir J. D. Hooker's paper, of a Cedar growing spon- 

 taneously in the island of Cyprus, with characters well-nigh intermediate between those of 

 the Lebanon and Mount Atlas forms. 



t The presence of an intermediate form between C. Libani and C. atlantica on the 

 island of Cyprus scarcely affects the general statement in the text ; the proximity of the 

 island to the Syrian coast almost brings it within the given range. 



