L o e ^v . The nectaiy and tlie sterile stanieu of Pentastemon etc. 87 



the 4z fertilo stamens all dilated at tlie base into broad 

 membranaceous spreadings fringed at the margin with 

 acute, stiff bristles. Honey is probably secreted by the 

 ground of the coroUa or elsowhere. Not any trace of 

 honey-glands at the base of the filaments is to be 

 fouud. Notwithstanding that the honey-protecting organs are 

 the most highly developed ones of the whole group. — Probably 

 visited equally by trochilids and insects. 



4. P. antirrJiiiwides Benth. — Spec. coli, by C. Gr. Pringle 

 on the San Diego Mts. (Calif.)! — 



Corolla up to 16 mm broad and at the middle inflated, the 

 sexual Organs strongly incurved and widely projected out of 

 the corolla as in some exploding ilowers, all 4 filaments 

 dilated at the base and set here by stiff hairs, but 

 without nectar-glands, higher up fitted with very minute 

 mieroscopical dentieles, the diso beneath the ovary plainly 

 developed , perhaps honey secreting. The staminodium is 

 7 — 8mm long, reaches the mouth of the corolla and has a 

 very striking form. Dilated at the base and bordered there 

 with acute bristles it is higher up reduced to an inferior thick- 

 ness and at the end enlarged again by a broad brush of thin 

 hairs. Its position in the mouth of the flower and its hairiness 

 seem to be significant of the mode of fertilisation worked pro- 

 bably by some strong insects of the family Apidae. 



5. P. Bothrockii A. Gr. — Specim. coli, by S. B, and 

 AV. F. Parish on the S. Jaeinto Mts. (Cahf.)! — Corolla very 

 small when compared with that of the foregoing 4 species, 

 bilabiate with a tube of 10mm and incurved stamens, all 4 

 filaments and the staminodium set a the base with 

 sliort stiff trichoms, but not dilated in the usual manner of 

 other Fruticosi, apparently without nectar-glands, the stamino- 

 dium naked, its point of Insertion very distant from the base 

 of the corolla, Its length is 7 — 8mm, so that it reaches with a 

 characteristic bending the mouth of the flower. — Probably a 

 bee-flower^). 



6. P. Lemmonl A. Grr. — Spec. coli, by M. E. Jones in 

 Nevada Co. (Calif.)! — The formation of the protecting organs 

 at the staminal base is the very same as in P. Rothrockii, 

 also the shape of the corolla, the position of the sexual organs 

 etc., but the staminodium carries a brush of long stiff hairs. — 

 Probably a bee-flower. 



As regards three other species belonging also to the 

 Fruticosi, namely: P. microphyllus A. Gr., corymbosus Benth. and 

 the mysterious P. fridescens Lamb. from Unalashka — my 

 materials were insufficient for any thorough study. The 6 de- 

 scribed species are in such a manner related that P. cordifoliiis 



1) As eil. Eobertson states, tlie flowers of P. laevigatus &n^ puhes- 

 cens are intended to be fertilised only by loug-tongued bees (see Flow. Jus. 

 Asclep. to Scropliul. Traus. St. Louis Acad. Sei. V : 590—591. 1881.) 



