Vor. I] STEWART—BOTANY OF THE GALAPAGOS ISLANDS 129 
BEMARLE Isu.: Villamil, on sand beaches and around brackish 
pools (no. 3197). BrinpLoE Ist.: Snodgrass and Heller. 
BRATTLE Ist.: (no. 3198). CHATHAM IsL.: Basso Point, on 
sand beaches (no. 3199); Wreck Bay, abundant near the 
beach (no. 3200). GARDNER IsL. (near Hoop Ist.) : on sand 
beaches (no. 3201). Hoop Ist.: Snodgrass and Heller. IN- 
DEFATIGABLE IsL.: southeast side, on sand beaches; north side, 
Snodgrass and Heller. JAmeEs Ist.: northeast side, common 
on sand beaches and on a rock one-half mile off the shore (no. 
3204). Srymour IsL., soutH: Snodgrass and Heller. Widely 
distributed. 
Hoindicum ©: Sp. Pl. 130).€1753); Kob: (1),. 192— 
CHARLES IsL.: in mud near a spring at 1000 ft. (no. 3208). 
CHATHAM Ist.: Wreck Bay, in shady places, 450-900 ft., (no. 
3206). Widely distributed in warm countries. 
H. parviforum L. Mant. 201 (1771); Rob. (1), 192.— 
ABINGDON IsL.: common on the lower parts, occasional above 
1000 ft., (no. 3209). ALBEMARLE IsL.: Cowley Bay, occa- 
sional at 1800 ft. (no. 3210) ; Iguana Cove, common near the 
shore (no. 3213); Tagus Cove, common to 1600 ft. (nos. 
3211-3212). Barrineton Ist.: Snodgrass and Heller. 
BraTT_e Iszt.: (no. 3214). CHampion Isz.: J. R. Slevin 
collector (no. 3215). CHartes Ist.: abundant in various 
situations all over the island (no. 3218). CHatTHAm Is-.: 
Wreck Bay, common near the shore and to 350 ft. (nos. 3216- 
3217). GARDNER IsL. (near Hoop Ist.) : (no. 3219). Hoop 
IsL.: occasional at 200 ft. (nos. 3220-3221). INDEFATIGABLE 
IsL.: northwest side, Andersson; southeast side, common on 
the lower parts and in the vicinity of the shore (no. 3222). 
James Isxt.: James Bay, abundant on the lower parts (no. 
3223). NaArporouGH Isx.: south side, Snodgrass and Heller. 
Tower Ist.: Snodgrass and Heller. \WENMAN IsL.: (no. 
3224). Widely distributed in warm countries. 
Tournefortia L. 
T. hirsutissima L. Sp. Pl. 140 (1753); Rob. (1), 193.— 
CuaTHAM Ist.: Chierchia. Robinson, |. c., suggests that this 
specimen may belong to Tournefortia rufo-sericea, a possibility 
which seems very likely, as subsequent collections have failed 
to show the species. Further distr. Mex., W. Ind., S. Am. 
