Bosn cftnina {Dog Sosp), 



3^18 NATURAL HISTORY OF PLANTS. 



orii^inally rescmblea tlic former. Tlie fruit (figs. 377, 378) is 

 iiuiltipk\ formed of a variable number of aclienes, enveloped in a 

 common sac or indusium, wliicli represents the floral receptacle, now 

 become fleshy throughout/ surmounted by the withered sepals or 



their cicatrices. Each achene is 

 glabrous, or part of its surface is 

 hairy.- Its walls are very hard 

 and thick,^ and surround a de- 

 scending: seed with membranous 

 coats, containing a fleshy embryo 

 with a superior radicle and elon- 

 gated cotyledons, touching by 

 their flat surfaces. There is no 

 albumen. 



The Roses are shrubs, erect, 

 branching, or creeping and climb- 

 ing. Most of them are covered with prickles of suberous nature,^ 

 scattered over the stems, the petioles, the veins of the leaves, and the 

 peduncles. Others are glabrous ; others, again, are covered with 

 glandular hairs. The leaves are alternate, imparipinnate, with the 

 leaflets often serrate ; and are provided with two broad membranous 

 stipules, adnate to the petiole for a great part of their extent, and 

 forming an incomplete sheath. In B. herherifolia^" of which it has been 

 proposed to make a distinct genus under the name of Hidfemia,^ the 

 leaves are reduced to a single leaflet, or perhaps rather to the base 

 of the petiole, on each side of which the stipules are much developed. 



Fig. 377. 



Fruit. 



Fig. 378. 

 Long, section of fruit. 



* The transformation into fleshy tissue may 

 even e.\tend to the pechnicuhir portion of the 

 floral axis ; in certain forms of R. alpina, for 

 instance, the summit of the pedunele is red and 

 succulent, like the iudusiuiu. The outer sur- 

 face of the latter often hears the hairs or prickles 

 which already existed in the flower, and may 

 now have increased in size. In the intervals 

 between the achcnes, too, the inner surface 

 bears the hairs whicli we have already described 

 on the disk (see p. 337, note 2). 



2 Kspecially on the two edges, more or less 

 projecting towards the centre, and the walls of 

 the receptacle, like those seen in Calycanthus. 

 Should one alone of these edges have hairs, it is 

 usually the one on the opposite side to the inser- 

 tion of the style. 



' Tlic mesocarp, which is quite dried up when 



ripe, is in some species fleshy and pretty thick 

 during nearly all the period of maturation. The 

 fruit is then rather a drupe ; we have made the 

 same observation as regards Calycanthece. 



•* They are formed by a hypertrophy of the 

 corky layer, here forming a large number of 

 projections with lenticular bases, the growth of 

 which produces no rupture in the epidermis, so 

 that this rises up over the whole of the prickle 

 to cover it with a tliin layer. 



* Pali., Nov. Act. Fetrop., x. 379, t. 10, fig. | 

 5._DC., Prodr., n. 25.— Ked. & TnoE., Ros., i 

 i. 27. — R. simpUc'ifuUa Sai.isu., Hart. Allert., 

 359 (ex LiNDL., Ros., 1) ; Pin: Lond. t. 101. 



^ DuMOET., Note sur Z'Hulthemia. — Endi., 

 Gen., n. 6358. — Lowea LiNDL., Rot. Re//., t. 

 1261.— Spach, Snit. a Riiffoii, ii. A^.—Rho- 

 dojt.ii.9 Ltcueb., Fl. Alt., ii. 224 (nee Endl.). 



