NOTES ON SPECIES OF LOTUS § PEDROSIA. 385 



Lotus arenarius Brot. This species is recorded from Palma 

 (Fueucaliente) in Phyt. Can. The only specimen which I have 

 seen is scrappy, and perhaps hardly sufficient for a satisfactory 

 identification if it stood alone. But in June, 1892, I found speci- 

 mens about the Combrecita (3800 ft.) in the same island, which 

 seem to me indistinguishable from certain Portuguese plants col- 

 lected near Cintra, which must, at least for the present, be placed 

 under L. arenarius, though from the description in the Frodromus 

 Florm HispanictE they should be very near L. canescens Kze., but as 

 I have never seen authentic examples of this plant, I am afraid to 

 speak more definitely. 



In his essay on Canarian botany (Engler's Bot. Jahrhilcher, ix. 

 1887) Dr. Christ has described a new species oi Lotus from Palma 

 (Barranco de Angustias, where it was collected by Hillebrand) 

 under the name of L. Hillebrandii. This he treats as intermediate 

 between L. spartioides and L. arenarius, quoting from Hillebrand's 

 notes : " Affinis L. spartioidei Webb et arenarii Brot. erectus" ; and 

 adding, after a long diagnosis, ^^ L. spartioides Webb Phyt. 81 differt 

 statura altiore diffusa, foliis miuoribus sericeis linearibus stipulis 

 conformibus, pedunculis teuuibus paucifloris, capitalis folio trifoliato 

 suffultis. L. arenarius Brot. dift'ert caule decumbente foliolis ob- 

 ovatis pube breviori, calycis dentibus lanceolatis vexillo breviore, 

 legumine minus nodoso." I have seen the types, and, except in 

 one particular, they seem to be identical with my Combrecita plant. 

 I can see no resemblance whatever to L. spartioides, and suspect 

 that neither Dr. Hillebrand nor Dr. Christ had much acquaintance 

 with that species, except from description, or from herbarium 

 "scraps." The erect habit ascribed to the plant may easily have 

 been accidental. There remains only the absence of a floral bract, 

 which is certainly curious. But it is a somewhat variable character 

 in other species of the genus. It is strange that Dr. Christ in the 

 same essay says of L. arenarius " capitulis floralibus aphyllis," 

 which as a general statement is certainly incorrect. I have there- 

 fore no hesitation in considering L. tlillebraudii as identical with 

 my Combrecita plant, which I believe to be a marked form or sub- 

 species of L. arenarius Brot. 



Lotus campylocladus Webb & Berth. A difficult and ap- 

 parently little kuown species, which seems to have been confused, 

 sometimes with L. holosericeus, sometimes with L. arenarius. I 

 suspect that it is confined to Tenerife, though there is a scrap from 

 Palma so named in herb. Webb. This probably belongs to L. 

 arenarius. 



The original description in Phyt. <'<ui. is as folloAVs : — " Jj. 

 villosus, foliis breviter potiolatis, foliolis anguslc obovato-ouneatis ; 

 stipulis ovatis petiolo subbrevioribns, calycc ultra medium diviso, 

 laeiniis lineari-lanceolatis, apicc subulatis, stylo 1-dentato." I 

 find, however, that the stipules equal or exceed the petiole, and are 

 subpetiolate. The habit is more or less prostrate or ascending; 

 the colour a curioas ashy green From L. holosericeus this species 

 may be readily distinguished by its much less erect habit, the 

 stipules unlike the leaflets, which are frequently retuse, and by the 



Journal of Botany. — Vol. 35. [Oct. 18i)7.] 2 c 



