30 



THE WEST-AMEEICAN SCIENTIST. 



men the accompanying label is the 

 following, viz. ; 



"No. 228 Guliemiu, 1830, 

 Gen. Nov. Donatia?Ex, Gay." 



Even as late as D. C. Prodro- 

 mus, this plant is included among 

 the "Dubus" at the close of Eriog- 

 onese, The natural inference from 

 the above ascertained facts is that 

 the obscure characters, and imper- 

 fect knowledge of this species, up 

 to a late, date prevented a clear 

 understanding of its true syste- 

 matic relations. With the publi- 

 cation of Gay's El. Chilensis, in 

 which this plant is figured, and 

 fully described under the name of 

 Lastarri?ea Chilensis, Gay, and the 

 subsequent elaboration of Prof. 

 Gray, and Bentham, its true place 

 in the natural system is clearly de- 

 termined. 



In venturing to differ from the 

 last named authorities as to its 

 generic character, in referring it to 

 a reduced form of Chorizanthe 

 (Proced. Dav. Acad. Science, Vol. 

 IT., page 47 ) . I vv r as forced to this 

 view in an endeavor to systematize 

 all the material in reference to this 

 genus, accumulated by extensive 

 field observation, and access to all 

 the avaiiabie collections and de- 

 scriptive works. 



The main point of difference 

 lies in the view taken of the floral 

 organs, and as to whether the in- 

 volucral appendage, or the peri- 

 anth, is obsolete in this particular 

 species. Noting, then, the marked 

 persistence of the involucral char- 

 acter ; on which, in fact, Bentham 

 mainly relied in establishing the 

 sub-order Eriogonese ; and seeing 

 no good and sufficient reason why 

 the organ in this species, which 



occupies the position and has all 

 the morphological characters of a 

 Chorizanthoid involucre, was not 

 such in fact. An explanation wa- 

 sought for the apparent absence 

 of the internal perianth by refers 

 ring it to a reduced form, as a lo- 

 bed ring in the throat of the invo- 

 lucre, its lower part being adnate 

 to the tube in a similar way, as the 

 stamineal whorl is not unfrequent- 

 ly adherent to the perianth, being 

 carried on its inner face as an ob- 

 scure thread, as well as occasion- 

 ally reduced in number, from nine, 

 to six, or three. What very mate- 

 rially helps to sustain this view,-is 

 the modification which the peri- 

 anth is plainly seen to undergo in 

 a regular series of reduction, be- 

 ing traceable from several flowers 

 to each involucre, pedicellate and 

 partly exsert, as in Centrostegia 

 section, gradually reduced to sin- 

 gle flowers by partial or entire ob- 

 soletion of the secondary ones, and 

 still farther in Euchorizanthe sec- 

 tion, becoming nearly or quite ses- 

 sile and strictly unifiorous, show- 

 ing also a more or less adnation of 

 the stamens to the tube of the peri- 

 anth, thus as it were foreshadoAv- 

 ing what we actually find in Cho- 

 rizanthe Lastarria3a,viz : a blending 

 of the perianth with the involu- 

 cral tube. As serving still farther 

 to confirm this view, it is possi- 

 ble by careful dissection of well 

 macerated young specimens to de- 

 tach the adherent ring, extending 

 as a membranous expansion lining 

 the involucral tube just as in the 

 analogous case of adherent sta- 

 mens, the filament can frequently 

 be separated as a continuous 

 thread to the base of the perianth. 



