179 



No. 306). The following note, taken from a letter of Mr. Brandegee, shows that his 

 specimens possess the characters of the genus as well as of this section : 



" In Abutilon cinctum there is no anthenferous column proper; each one of the car- 

 pels is transversely septate, the upper part hearing two collateral seeds and the 

 lower cell one. Very little of it was collected, and I can give you only the fragments 

 inclosed in this letter. " 



Wissadula tricarpellata Robinson & Greenman, sp. nov. 



Stems terete, lignescent below, finely tomentose-pubescent, with short simple (vis- 

 cidulous?) hairs and somewhat scabrous; the surface in age Blit with numerous fine 

 longitudinal linear depressions, apparently the result of the spreading of the cortical 

 fibers; leaves ovate, gradually narrowod to a caudate acuminatum, serrate-dentate, 

 green and finely stellate-pubescent on both surfaces, slightly paler beneath, usually 

 marked above with a small dark-purple spot at the origin of the (5 to 7) nerves, 

 cordate at the base, 6 to 10 cm. long, 3.5 to 5 cm. broad ; petioles 3 to 7 cm. long, 

 tomentose; stipules and bracts filiform, pubescent, caducous; flowers numerous, 

 borne in a large, loose, racemose panicle; pedicels 0.6 to 2 cm. long; calyx canipanu- 

 late, stellate-tomentose upon the outer surface ; segments ovate-lanceolate, attenuate, 

 4 to 5 mm. long, somewhat exceeding the tube; petals obovate, cuueate, essentially 

 entire, 7 mm. long, densely stellate-pubescent near the base upon the outer surface 

 and margins, in dried specimens bluish; stamineal tube 1.7 mm. long; style deeply 

 3-parted; divisions puhescent below; ovary and capsule tomentose; ovules 3 in 

 each cell; seeds dark brown, 2 mm. in length, inconspicuously verrucose.— Collected 

 by C. G. Pringle on a moist hillside at Tequila, .Jalisco, October 17, 1893, No. 4578 (type 

 number), and September 30, No. 4610 in part. Both numbers were distributed with 

 Mr. Pringle's sets of 1893. Unfortunately, the material under No. 4610 seems to 

 have been confused and to consist in some sets of a Sida near to if not identical 

 with »S'. alamosana Wats. This material may be distinguished from the Wissadula, 

 here described, not only by the well-known carpellary difference between the genera 

 Sida and Wissadula, but by the pubescence of the leaves, which is longer and chiefly 

 simple. 



"Wissadula wissaduloides (Baker f.) Rose; Abutilon ivismdirfoides Baker f. Contr. 

 Nat. Herb. 3: 312. 1895. 



The type of this species is in the National Herbarium (type No. 482). The species 

 is only known from Dr. Palmer's specimens from Ymala, State of Sinaloa, 1891 (No. 

 1720). 



MISCELLANEOUS SPECIES. 



Bombax palmeri Wats. Proc. Am. Acad. 22 : 399. 1887. 



A tree 4.5 meters or more high, with a large spreading top; bark very scaly, but 

 finally falling, leaving a smooth, reddish trunk; leaflets mostly 5, clothed when 

 young with a dense stellate tomentum on both sides, becoming nearly glabrate 

 above in age; petals 5, 10 cm. or more long, t.hickish, densely stellate; stamens 

 numerous, united at base into a tube 2 cm. long. 



This species has heretofore been known only from Palmer's and Pringle's collec- 

 tions about Guadalajara. It proves to be one of the commonest trees in the trop- 

 ical district from Rosario to Acaponeta and Tepic, eastward to the mountains, and in 

 the tropical valleys of the interior. The trees are usually found on rocky hillsides 

 at from 200 feet to 3,600 feet altitude. It flowers some time during the dry season 

 (October to June), fruiting just before the beginning of the rainy season. Flowers 

 heretofore have not been collected, and mine are poor, consisting only of old and 

 withered ones. 



Specimens were collected near Acaponeta, Tepic, June 23, 1897 (No. 1451), and 

 near Colomas, Sinaloa, July 21 (No. 3215). Trees were noted on the east side of the 

 west range of the Sierra Madre at an altitude of 3,600 feet. 



