66 CRUCIFERAE 



Tendency to li-af dimorpliisni. — Tlie basal leaf rosettes, especially those borne on sterile 

 shoots, commonly exhibit dense silvery tufts of very narrow entire leaves. In the most marked 

 state of the species the broader caulinc leaves are pectinately incised. The differences between 

 the two extreme types of leaves show, therefore, a strong development in the direction of dimor- 

 phism. In addition to its peculiar foliage the species is distinguished by the attenuate or some- 

 what gladiate form of the pods. 



Locs. — Shackelford Canon, Butler 17C3 (in part) ; Greenhorn Mt., Butler 685 (whito- 

 flowcred) ; Humbug Mt., Butler 587; Yreka, Butler 587, 1446; Edgewood, J. W. Eisling. Ore.: 

 Tennessee Pass near Kirby, Henderson 5918. 



Refs. — AR.^B1S subpinnatipida Wats. Proc. Am. Acad. 20:353 (1885), based on spms. from 

 West Humboldt Mts., Nev., Watson, Scott Valley, Siskiyou Co., Greene, and Waldo, Ore., Howell; 

 Jepson, Man. 431 (1925). 



12. A. secunda Howell. Summit Rock-cress. Biennial or perennial; stems 

 simple, several from the base or solitary, branched above, 8 to 18 inches high ; herb- 

 age densely pubescent below, scantily pubescent above; leaves 1 to IV2 inches long, 

 the blades of the basal petioled, oblanceolate, repand-toothed, the blades of the 

 cauline sagittate-clasping, oblong-linear, repand or entire; flowers white, becom- 

 ing pendent on pubescent pedicels; petals white, 3^^ lines long; pods narrow, se- 

 cund, deflexed, 1^ to 2^2 inches long. 



Montane slopes, 5000 to 6000 feet : northern Humboldt Co. and western Siski- 

 you Co. North to Washington. June-July. 



Tax. note. — Certain American authors, particularly the authors of the Botany of California 

 and of the Synoptical Flora of North America, have not infrequently determined certain West 

 American plants of the Transition Zone as Arabis holboellii Ilornem. This species was based 

 on specimens from Disco Island in the Arctic Ocean and is portrayed in an excellent illustration 

 in the Flora Danica, pi. 1879. While the present writer has also (Man. 429, — 1925) referred 

 various Californian plants to this species in the status of varieties, such reference of California 

 plants, which are in some cases of the Arid Transition Zone, to a truly boreal species seems so 

 problematical that the step is here taken of recognizing these plants of the Arid Transition under 

 specific designations. 



As figured in the detail drawing in the Flora Danica (vol. 11, pi. 1879) the petals of A. hol- 

 boellii are 2^4 times as long as the sepals; the flowers and fruits are both distinctly secund; the 

 flowering raceme is loose; the flowers spread horizontally; the petals are white; the filaments are 

 dilated at the base and middle and contracted above; and the pods are (for the most part) only 

 slightly curved. Of our Californian forms A. secunda Howell is, perhaps, nearest the plant of 

 the Arctic Ocean, but its sepals are a little over half as long as the petals; its raceme is at first 

 capitate-congested, though promptly becoming looser below with the flowers soon pendulous; its 

 longer stamens have filaments equally dilated throughout. 



A. retrofracta Graham is often quoted as a synonym of A. holboellii; as represented by 

 Californian material, A. retrofracta has a coarser and more abundant pubescence; its racemes 

 are not secund; its pedicels are usually pilose. A. arcuata Gray is even further removed; its 

 habit is coarser; its corolla is usually dark red; its petals are about 1^/^ times as long as the 

 sepals ; its racemes are not secund, but its pods are strongly curved. 



It appears that a series of allied forms reaches across the continent from Southern California, 

 Arizona and New Mexico to Disco Island in the Arctic Ocean, the type locality for Arabis hol- 

 boellii. While the northerly end of the series is not well known to us and much is still lacking in 

 the way of knowledge of the life history of our own forms, we are disposed, at this time, to look 

 upon certain of the published segregates of the Arid Transition Zone as best disposed in specific 

 rank. There seems to be fairly sufiicient grounds, geographically and morphologically, to warrant 

 this step. 



In Pittonia (4:187-189) E. L. Greene has, to be sure, already taken the stand that Arabis 

 holboellii is to be excluded from the North American mainland flora and he publishes segregates 

 formerly resting under the larger concept, but he docs not organize his series monographically. 

 Consequently one can not so readily make appraisement of his species or the basis of his ideas. 

 While his diagnoses bespeak the polished scholar, his usual practice led him to quote but a single 

 specimen or only a few specimens, so that, sometimes, his species are, perhaps, neglected from 

 lack of means to appreciate them. For example, after much study, Arabis rectissiraa Greene is 

 now, apparently, reasonably clear to us. Regardless of what differences of opinion may even- 

 tually develop as to its status, it is a fairly definite form of the Sierra Transition Zone, specimens 

 of which, derived from the length of that mountain chain, can be quoted freely from a number of 

 large American herbaria. 



Some of our units of this group are still quite diverse in content. Arabis retrofracta Gra- 

 ham is one of these. As received here the plants included under this name are varied in appear- 



