PIPER—BERBERIS AQUIFOLIUM AND BERBERIS REPENS. 445 
given: ‘Pursh having erroneously added the flowers of B. aquifolium to his 
plate of nervosa, misled De Candolle, who has reproduced the species under the 
name of glumacea.’”’ 
Dr. Gray evidently refers to the Menzies specimen then in the 
Lambert Herbarium now at Kew. The comment in the British 
Flower Garden quoted above is seemingly not the reference Dr. Gray 
intended, and it can scarcely be the remarks of Sweet previously 
quoted. 
Inasmuch as the first part of Torrey and Gray’s Flora of North 
America containing their treatment of Berberis was published in 
1838, these notes of Dr. Gray could have had nothing to do with the 
conclusions reached by them at that time. 
In Dr. Gray’s last publication on the subject *® he writes under 
Berberis aquifolium Pursh, “ Fl. 1: 219, in part and ¢. 4, mainly,” 
while under Berberis repens Lindl. he states, “B. aquifolium Pursh, 
]. c. 219, mainly as to descr.” This apparently means that Pursh’s 
plate is mainly the shiny-leafed tall plant and his description mainly 
the dull-leafed low plant, but the basis for these conclusions does 
not appear. | 
The fundamental error of Lindley, as likewise of Nuttall before 
him and of later authors who have followed them, lies in the assump- 
tion that the seeds brought back by Lewis were of the same plant of 
which he collected specimens in flower at the Great Rapids of the 
Columbia. When Lewis was at the mouth of the Columbia he ex- 
pressly notes that he had not seen either the flowers or fruit of the 
“mountain holley.” He recrossed the Bitter Root Mountains in the 
latter part of June, 1806, far too early to have secured ripe fruit in 
the neighborhood of Kamiah and Weippe, Idaho, where he had been 
during most of June. It is apparently certain, therefore, that he 
secured the seeds he brought back east of the Bitter Root Mountains 
and most probably in Montana. Nuttall says “Rocky Mountains,” 
but in Lewis’s journal no record of the collecting of these seeds has 
been found. It is certain that he could not have gotten the seeds at 
the Great Rapids, where he collected the types of B. aquifolium and 
B. nervosa in flower. , 
This brings us to the question as to whether the type of B. aqui- 
folium Pursh, collected at the Great Rapids, is the same species .as 
B. repens Lindl., grown from seed collected by Lewis probably in 
Montana, where only this latter species occurs. 
The writer has previously expressed the opinion * that this could 
not be the case, as the glaucous-leafed species, B. repens, was not 
known to occur so far down the Columbia River as the Cascades. 
* Syn. Fl. 1: 69-70. 1895. *™ Contr. U. S. Nat. Herb. 11: 282, 1906. 
