CliASSTJLACEM. 



317 



of Crassulacea, has neither the fleshy leaves, the habit, nor the dehis- 

 cence of the fruit of this order ; it links it to Saxifragacece, wherein 

 we class it. Its seeds have albumen, though far from copious ; 

 and this is really absent 1 as a rule in the ripe seeds of true Crassu- 

 lacea? When Saxifragacece have their carpels free, they are thereby 

 distinguished from Crassulacece, which moreover are never woody. 

 The Francocce, which among Saxifragacece come nearest to this order 

 after Pent/iorum, are distinguished by the presence of albumen and 

 the more intimate union of the carpels. The free gland found in 

 Crassulacece outside the base of each carpel, is also usually absent in 

 Saxifragacece? De Candolle has also shown how Crassulacece pass 

 into Illeccbrece through Tillcea. J. Gr. Agardh 4 also allies this order 

 to Tamariscinea. We ourselves, while holding like most living 

 authors that it is inseparable from Saxifragacece, find that it is also 

 closely allied to Astrocarpece, the polycarpic type of Resedacece. In 

 Astrocarpus the vegetative organs are comparable with those of 

 Penthorum ; the gynseceum consists of free carpels as in the House- 

 leeks. And as the carpels become pluriovulate in Resedece, the only 

 difference left between the two groups lies in the irregularity of 

 corolla, androceum, and disk in Resedacece. 



Many Crassulacece owe their uses 5 to the quantity of water contained 

 in their fleshy succulent organs; this is especially true of the House- 

 leeks (Joubarbes), which are used as cooling applications to burns, 

 abscesses and haemorrhoids, and to soften warts and corns. The leaves 

 are used entire, sliced or bruised, raw or rarely cooked, alone or in 

 combination with oil or grease. The common Houseleek, 6 which is 



1 A. L. de Jussiett admitted its presence: 

 " utpote perispermo seminis farinaceo instruct ce ;" 

 an opinion shared by Bentham & Hookeb {Gen., 

 656) : " albumen carnosum." Exdlicueb writes 

 (Gen., SOS, ord, 169) : "embryo in axi albuminis 

 orthotropus," and Lindley (Veg. Kingd., 344, 

 ord. 120) : "Embryo straight in the axis of fleshy 

 albumen:' But the existence of albumen is rightly 

 denied by other authors (Ad. Be., Eiutm., xxviii. 

 cl. li. ; — Lem. & Decne., Traite Gen., 237). 



2 " Cephalotus is another type closely linking 

 Crassulads and Saxifragads." (Adansonia, 

 vi. 3.) 



3 " A Crassulaceis discrepant (Saxifragacece) 

 praecipue habitu, defectu glandularum hypogyn- 

 arum carpellisque rarius polyspermis ; ted Crassit- 

 lacea potius subordinem Saxifragaceamm quam 

 ordinem sistunt." (B. H., Gen., 629.) 



4 TJisor. Syst., 343, t. 25, figs. 6-9. " Crassu- 

 lacea sunt Tamariscinea} herbaceae et nobilius 

 corollata?, mecliante pra;cipue Telephio versus 

 Curvembryas tendi'ntes." 



5 Endl., Eiichirid. ,4>0(). — Linde., Veg. Kingd., 

 345; Fl. Med., 275. — Gdib., Drog. Simpl., ed. 6, 

 iii. 254. — Rosenth., Syn. PI. Biaplior., 573. 



6 Semptrviium tectorum L., Spec, 664. — DC, 



