ROSACEA. 



3o7 



multiple fruit, and each achene encloses a seed, containing a fleshy 

 exalbuminous embryo, with its radicle superior. 



The Strawberries are perennial herbs ; the stem is a short sym- 

 podium,' and the leaves are alternate, trifoliolate, digitate, or rarely 

 pinnate, with two lateral petiolary stipules. The branches are often 

 prolonged into runners with scattered leaves, whose axillary buds 

 stril^e root in contact with the soil" (fig. 413). The flowers are 

 terminal, solitary, or more frequently collected into alternate, few- 

 flowered, often uniparous cymes at the summit of a conmion 

 peduncle. A large number of species have been described, inhabiting 

 all the temperate and alpine regions of the northern hemisphere, and 

 the mountains of South America and the Mascarene Islands.^ But 

 most of these are only forms or varieties, and there are probably not 

 half a dozen true species. 



The Potentils' (figs. 420-427) come very near to the Strawberries 

 in their perianth and androceum ;' and the true species are only dis- 

 tinguished by two characters, which are sometimes very ill-marked.** 

 The style is usually inserted higher up on the ovary, and so the 



similar consistency to tbat of Comarum. Wlien 

 it is not very fleshy the achenes may sepiirate 

 from it when quite ripe, as in certain Potentils. 



1 IHM. (r.), Bot. Zvit., viii. 250.— Wyld., 

 Flora, xxxiv. 364.— Gben., JBuU. Soc. Bot. de 

 IV., ii. 349. — J. Gay, Ann. Sc. Nat., ser. 4, viii. 

 185. 



2 A. S. H., Morph. Veg., 235.— A. Juss., 

 Mem., 156.— Patue, Elem., 58, fig. 93. 



3 DucH., Eist. ^\lt. des Frais., 1766.— 

 Feenz., Frag., 1662. — L., Fraga resca, 1772. 

 —DC, Prodr., ii. 569.— LouK., Fl. Cock., 325. 

 — RoxB., Fl. Ltd., ii. 520. — Wight & Abn., 

 Prodr., Fl. Pen. Ind., i. 300. — Wight, Icon., 

 t. 988, 989.— MiQ., Fl. Ind. Bat., i. p. i. 371. 

 — H. B. K., Nov. Gen. et Spec, vi. 172.— C. 

 Gat, fl Chil, ii. 315.— Tore. & Ge., Fl. N. 

 Amer., i. 447. — A. Geat, Man. of Bot., ed. v. 



155 J. Gat, loc cit., 194. — W alp.. Rep., ii. 



25 ; Ann., i. 277. 



* Potentilla T., Inst., 295, t. 153.— L., Gen., 

 n. 634.— J„ Geti., 338, 453.— G^etn., Frvct., 

 i. 350, t. 73. — Lamk., Diet., ii. 527 ; Suppl., ii. 

 667, III., t. 442.— Nestl., Mon. Potent., 1816. 

 — Lehm., Mon. Potent., 1820-35. — DC, 

 Prodr., ii. 571.— Spach, Suit, a BuJJon, i. 409. 

 — Endl., Gen., n. 6363.— B. H., Gen., 620, n. 

 48.— Quinquefolium T., Inst., 296, t. 153.— 

 Pentaphylloides T., op. cit., 2l>8. — Adans., 

 Fam. des PI. ii. 295.— G^etx., Fnict., i. 349, 

 t. 73. — Fragariastrum SCHKUE, Fnum. Plant. 

 Transyhan., 137.— Bcolia Big., Fl. Best., 351. 



* The number of stamens is here as variable 

 as in Fragaria. A. Dickson, who has studied 

 the arrangement of the stamens in Rosacecs 

 generally very fully (^ee Journ. of Bot., iii. 

 (1865), 209), and contiruied most of the results 

 obtained by Patee on this subject by organogenic 

 study (see p. 337, note 1), has especially deter- 

 mined the number of pieces in the audroceuui 

 and their arrangement in Potentillas {On the 

 Staminal Arr. in some Spec, (f Potent., and in 

 Nuttallia cerasiformis ; Jcurn. of Bot. iv. (1806), 

 t. Iii.). He has sliown that in certain species, 

 such as P.Jruiicoia, the andrcceum furnis five 

 festoons, each cor.taining four or five stamens, and 

 extending from petal to petal; the convexity of the 

 festoons is towards the centre of the flower, and 

 there are no stamens superposed to the sepal:-. 

 On these grounds the author considers the andro- 

 ceum as formed of five compound stamens, the 

 terminal lobe of each being developed as a petal 

 so called, and the lateral lobes as fertile stamens. 

 In other species, where he finds a stamen exactly 

 superposed to a sepal, he considers it the repre- 

 sentative in the androceum of one of the caly- 

 cular leaves, which are of stipular i;alure, and 

 hence alternate with the sei als just as the 

 opposilisepalous stamens alternate with the op- 

 positipetalous staminal bundles. 



6 So ill-marked indeed that we should cer- 

 tainly be consistent in refusing to retain Fra- 

 garia and Potentilla as distinct genera. 



