99 



ization and placed it in his book and described it as intermediate 

 between A. caiiadensis and a species of American Ainela}ic]iier 

 which he called ''A. ovalis {^Crataegus spicata Lam.)." 



A dwarf form of this class grows on the rocks at Bellows Falls, 

 Vt., between the two railroad bridges. It is so interesting that 

 it is proposed here as a new species : 



Amelanchier saxatilis sp. nov. Small, spreading, partly 

 prostrate shrubs, i to 3 feet high, irregularly and abundantly 

 branched ; branches slender, twigs very slender ; breaking buds 

 slim, rather woolly, scales narrow ; bracts rose-color. Leaves 

 very woolly on the lower side when young, glabrous when full- 

 grown, oval, rounded at the base, points short and broad, finely 

 serrate, 1.5 in. long by i in. wide. Flowers early, abundant, 

 rather small, well proportioned, 0.75 in. broad, very white, in 

 very numerous, spreading, slender racemes ; peduncle, pedicels, 

 and calyx densely woolly. Petals very narrow, three times as 

 long as wide. Fruit small, ripening early, in slender racemes, 

 top of pome glabrous. 



Type in the herbarium of W. H. Blanchard. This form has 

 much of the habit of the sand cherry, Priiiuts puinila L. and 

 when in flower would be taken for it at a little distance. This 

 delicate species is not the dwarf form of A. intermedia Spach, 

 which grows on the rocks near the coast, often hardly a foot 

 high, erect and bearing fruit. 



A. OLiGOCARP.\ (Michx.) Roem. The mountain form with 

 {&\N flowers has generally had but one name, though sometimes 

 it has been called A. sangiiinea, which name belongs to another 

 species. It is quite variable and has been little studied. Dr. 

 Britton has recently segregated from it 



A. ARfeuTA Nutt., which is perhaps more correctly A. argiita 

 Britton, the name, it appears, having been given first to an her- 

 barium specimen by Nuttall but first pubHshed in Britton's 

 Manual, ed. 2, p. 1066, without citation of a type or any def- 

 inite reference to the source of the name. The species, however 

 is more fully described in Torreya 5: 107. Je 1905, by Mr. 

 Eggleston, who indicates the materials from which the original 

 description was drawn. 



The forms of the second class which have been called sangidnea, 



