SAXIFRAGE FAMILY 155 



19. R. menziesii Pursh. Canon Gooseberry. Openly branched bush 3 to 8 

 feet high; stems more or less densely prickly, with mostly 3 spines at the nodes; 

 bark brownish; leaf -blades 3 to 5-cleft, toothed, truncatish at base, glabrous or 

 nearly so above, finely pubescent beneath and also with scattered gland-tipped 

 bristles, % to 1 (or IV2) inch wide, the petioles about 3 to 6 lines long; peduncles 

 1 or 2-flowered, 10 to 12 lines long, the 2 bracts discrete, borne above the middle; 

 flowers 6 to 8 lines long; calyx red-purple, a little puberulent and with a few mi- 

 nute scattered gland-tipped bristles; calyx-tube 1 to 1^4 lines long; filaments 

 equaling to 1% times as long as the white petals; anthers shortly but distinctly 

 apiculate; style exceeding the stamens; ovarj^ globose, rather thickly covered with 

 gland-tipped spines of unequal length and exhibiting a little whitish puberulence ; 

 berry spiny, 5 to 6 lines in diameter, the spines 1 14 to 2 lines long. 



Canons and flats along the coast, 50 to 700 feet : Humboldt Co. to Del Norte 

 Co. North to southern Oregon. Feb.-Apr. 



Locs. — Van Duzen gravel bar near Carlotta, Tracy 5985; Eureka, Tracy 2941, 2130; Blue 

 Lake, Humboldt Co., Tracy 35G8; Drakes Hill near Alton, Tracy 5421; Luffenholtz Creek, Trini- 

 dad, Tracy 4866 ; Crescent City, Howell 1432. 



Endemic variants. — Eibes menziesii, in a broad sense, is a variable species of the Eed- 

 wood belt, and thence to Los Angeles Co., but always in the immediate vicinity of the coast, 

 save for the var. hystriculum of Mt. Diablo, the var. leptosmum, which recurs at Mt. Diablo, 

 and for the var. hesperium, the most pronounced southern variation, which extends somewhat 

 inland. The original specimens of the species were collected at Trinidad on the Humboldt coast 

 in 1791 by Archibald Menzies of the Vancouver Expedition and are well matched by modern 

 specimens collected at Eureka (Tracy 2941). In an extremely narrow and restricted sense this 

 original botanical type occurs only on the Humboldt coast, but is everywhere subject to slight 

 modifications as it ranges southward. These modifications are, with a few exceptions, too trivial 

 to merit even varietal recognition, although some of them are by us thus ranked because of a 

 certain circumstance in relation to localized differentiation. Each marked or definite minor 

 topographic and climatic area along the coast is characterized by a Eibes menziesii form bearing 

 a certain combination of structural features, and this plant form occurs only in its ovra particular 

 area, representing in each case extremely refined differences in pubescence or distribution of 

 pubescence or absence of such ; in degree or character or distribution of glandulosity or absence 

 of such; and in size, length or number of the berry spines. While these features are so slight 

 as to defy in all probability satisfactory written diagnosis, yet, after familiarity born of years 

 of study in the field, garden and herbarium, we are able by aid merely of a lens to identify each 

 particular geographic form. For example, the California specimens of David Douglas (collected 

 in 1831) never carried a special locality, but one may now say of his specimens that Eibes men- 

 ziesii var. hystrix Jepson was collected by him with scarcely any doubt at Point Lobos near Car- 

 mel, and not further southward on the Monterey coast (where the var. hystrix also occurs), and, 

 similarly, that his specimens of the var. hesperium Jepson were collected at Santa Barbara, and 

 not easterly in the San Gabriel Mountains. These minute differences thus come to have a special 

 significance, because it would appear that they are inherited in each district over long periods 

 without appreciable variation. This species is, therefore, interesting for its minute geographic 

 variants in association with minor climatic districts along the Californian coast. All these 

 geographic phases of the species would be explained by some botanists as hybrid segregations 

 but we prefer to think that they are climatic and topographic variants, since each is so closely 

 associated with the climatic elements of rainfall, temperature, humidity and insolation, and the 

 edaphic factors. 



Tax. note. — We find that relative length of petals and stamen-filaments is unimportant as a 

 distinguishing mark in the Eibes menziesii group, but significance may be attached to the char- 

 acter of the fruit and, within certain limits, to pubescence and glandulosity. The type of Eibes 

 leptosmum Gov. has been compared by us with the type of Eibes subvestitum H. & A. at the Eoyal 

 Botanic Gardens, Kew, and found to be identical. The type of E. subvestitum H. & A. is, in 

 turn, essentially identical with the type of E. menziesii Pursh, preserved at the Natural History 

 (British) Museum, differing only slightly in one particular, that is in a minute degree of glandu- 

 losity of the lower side of the leaf. This extremely slight difference, however, represents a link 

 in the chain of variants southward along the coast and we, therefore, on this account, retain the 

 name var. leptosmum, thus completing a geographic series, which is to be described as follows : 



Var. leptosmum Jepson. Similar to the species, the leaves glandular-dotted (or vsdth sessile 

 glands) beneath as well as with stalked glands, but somewhat less puberulent. — Low hills and 

 canons, 10 to 900 feet: flats or hillslopes, Mendocino Co. to Santa Clara Co.: Ft. Bragg, W. C. 

 Mathews 99; Inverness, Marin Co., Jepson 503; Olema, Marin Co. Jepson 13,512; Mt. Tamalpais, 



