GILIA FAMILY 189 



lent; leaves V2 to 2 inches long, the blades pinnately divided into discrete linear 

 segments, the segments entire or with 1 or 2 teeth or lobes, 1 to 3 lines long ; upper 

 leaves with coarser and fewer divisions ; flowers 2 to 5 in close short-pedicelled ter- 

 minal clusters or sometimes solitary and terminal; calyx 3 lines long, its teeth lan- 

 ceolate ; corolla blue, tubular or sliglitly expanded upward, 3V2 to 4 lines long, its 

 tube and throat shorter than or not exceeding the calyx; corolla-throat with dark 

 purple spots ; fruiting calj^x conspicuously accrescent, 4 to 6 lines long, its lobes 

 broadly short-lanceolate; capsules large (3 to 4 lines long). 



Ocean sand dunes or sandy soil, 5 to 40 feet : along the coast line from Del Norte 

 Co. to Sonoma Co. May-June. 



History of the species. — The binominal Gilia inconspicua Sweet rests directly on Ipomopsis 

 inconspicua Sm., a species based on a garden plant ^rown in the year 1793 at Sion House, on the 

 north bank of the Thames River opposite the Roj-al Botanic Gardens at Kew, England. The plant 

 was raised from seed thought to have come from "America." No herbarium specimens were pre- 

 served, but the describer of the species, J. E. Smitli, illustrated the diagnosis with a colored draw- 

 ing in his Exotic Botany (vol. 1, t. 14, — 180i). Notwithstanding the fairly extended description 

 and the illustration, the identity of Ipomopsis inconspicua has been doubtful. Traditionally it has 

 been regarded as the same species as Gilia inconspicua Dougl., a small-flowered Gilia published by 

 W. J. Hooker in 1829 on the basis of a collection made by David Douglas on the "sandy barren 

 on the southern branches of the river Columbia on the northwest coast of America, growing under 

 the shade of Purshia tridentata and some species of Artemisia." Ipomopsis inconspicua Sm. 

 could not, however, have been discovered in the interior of Oregon and Washington, because at the 

 time of its cultivation in England no white man had ever traversed that region, and Gilia incon- 

 spicua Dougl. does not grow west of the Cascade Mountains. Moreover, the various forms of Gilia 

 inconspicua Dougl. in California, here described under the name Gilia tenuiflora Benth. var. sinuata 

 (Dougl.), do not correspond well to the description and illustration of Ipomopsis inconspicua. 



The origin of the seed stirs curiosity, since in that early day it must have been derived from 

 some coastal region, rather than an unknown and untraveled interior. Although few of them car- 

 ried scientific explorers, a good many ships had visited the coast of California in the eighteenth 

 century. Therefore, in the sense of possible physical access, the seed, to be sure, could have come 

 from the Coast Ranges of California and it is well to consider more curiously this possibility. 



In the South Coast Ranges the species Gilia multicaulis Benth. is widely distributed and often 

 occurs near the coast line. GUia inconspicua Sw. could not be this species, however, for the follow- 

 ing reasons: 1. The stems and branches of Gilia multicaulis are more slender than in Ipomopsis 

 inconspicua. 2. The foliage of Gilia multicaulis is finer or more finely divided than in Ipomopsis 

 inconspicua. 3. The peduncles in Gilia multicaulis are more slender tlian in Ipomopsis inconspicua 

 and usually with more flowers in the terminal cluster. One of the peduncles in the illustration of 

 Ipomopsis inconspicua is rather short, rigid and 1-flowered. This type of peduncle is exactly that 

 which is sometimes found in Gilia millefoliata of the north coast line. The peduncles of Gilia 

 multicaulis var. peduncularis (Eastw.) are not infrequently 1-flowered, as sometimes in I. incon- 

 spicua, but the peduncles are usually long and slender or even filiform, whereas they are short and 

 stout in I. inconspicua. 4. The inflorescence in Gilia multicaulis tends to be naked, that in I. incon- 

 spicua tends to be leafy. 5. The corolla in the type specimen of Gilia multicaulis has a slender tube 

 with a broadly expanded throat. The corolla in Ipomopsis inconspicua is broadly tubular (and 

 approximates the corolla of Gilia millefoliata of the North Coast Ranges). 6. In Gilia multicaulis, 

 the corolla is H4 to 3 times as long as the calj'x. In Ipomopsis inconspicua, the corolla only slightly 

 exceeds the calyx; it is, as it were, crowded dovra into the calyx (a condition which recalls that in 

 Gilia millefoliata of the North Coast Ranges). 



The region of the North Coast Ranges may next be considered. In 1838 Fischer and Meyer 

 published a small-flowered Gilia from Bodega in Sonoma Co. as Gilia millefoliata. On the basis 

 of the description, reprinted in Linnaea (13: Litt.-Bericht. 109, — 1839) and in Erythea (2:164, — 

 1894), and a topotype from Bodega, this is an ocean sand dune plant, usually rather low, with 

 stoutish branching stems, flowers few in rather close glomerules or sometimes solitary, the corollas 

 little exceeding the calyx and corolla-tube and throat not at all exceeding it, and, of most critical 

 importance, the corolla-tube and throat tubular or only slightly expanded upwards. Nearly all, 

 perhaps all, Gilias whose flowers are borne in close clusters or even dense heads, develop occasion- 

 ally or frequently loosely-flowered or openly cymose or paniculate forms. Gilia millefoliata some- 

 times occurs in loosely-flowered inflorescences. Most interesting in this connection is the fact that 

 a loose-flowered specimen of this species, which has been preserved, was grown in a garden by 

 J. P. Tracy. On the whole, then, it is believed that a loosely flowered form of Gilia millefoliata 

 corresponds more closely to the illustration and description of Ipomopsis inconspicua Sm. than 

 any other available material. 



Arguments for the decision that Ipomopsis inconspicua Sm., and not Gilia inconspicua Dougl., 

 is the equivalent of Gilia millefoliata F. & M., or a form of it, are here summarized as follows: 

 1. Ipomopsis inconspicua Sm. is a leafy-stemmed plant, whereas Gilia inconspicua Dougl. in its 



