RvDBEKG : Notes on Piiilotria Raf. 4o9 



as described. He stated also that he had examined eight specimens 

 collected by Moser in Pennsylvania, in 1832, and one by Schwein- 

 itz. These were all hermaphrodite, with from 2 to 7 stamens. 

 In the Torrey herbarium there is a specimen of Moser's collection, 

 but the tops of the flowers are broken off. In this specimen, how- 

 ever, the leaves are not oblong and obtuse, as described by Michaux, 

 but linear and acute. There are other specimens in our herbaria 

 with oblong or oblong-ovate leaves. One of these represents 

 Udora cauadcnsis of Torrey's Flora of New York from Torrey's 

 sets distributed in connection with that publication. The flowers 

 are hermaphrodite, as Torrey described his fertile flowers. Torrey 

 stated that he had not seen any living sterile flowers but drew 

 his description of them from material collected by Engelmann. 

 This material I shall discuss later. There are other broad-leaved 

 specimens, viz. : from Watertown, N. Y., 1834, Dr. Gray ; East 

 Haven Pond, Conn, (collector not given) ; Thousand Islands, N. 

 Y., July 16, id>^g, Jolm Northrop ; Prior's Lake, Minn., 1891, C. 

 A. Ballard ; Fish Creek, Sylvan Beach, N. Y., L. M. Underwood 

 321 1 ; Waynesborough, Va., 1897, W. A. Miirrill ; Chilson Lake, 

 N. Y., 1900, Dr. & Mrs. N. L. Britton ; but all these have no 

 flowers or the flowers are in such condition that it is impossible 

 to say whether they are pistillate or hermaphrodite. None of 

 these specimens has any sessile staminate flowers. Then the 

 question arises : is the broad-leaved, /. e., the typical Phdotria 

 canadensis (Michx.) Britton always hermaphrodite ? Nobody, so 

 far as I can find, has described any staminate flowers in connection 

 with broad, oblong leaves. 



Serpicula verticillata Muhl. Cat. PI. Am. Sept. 84. 18 13. 



This is described as monoecious and 3-androus, but the state- 

 ment that the sepals and petals of the staminate flowers are four 

 must have been an error or else Muhlenberg's specimens were 

 abnormal. What it really was is of little consequence, as it is not 

 Serpictda ver tied lata L. 



Serpicula verticillata angustifolia Muhl. Cat. PI. Am. Sept. 



84. 1813. 



This is of interest as it is the first name as far as known appli- 

 cable to a form evidently distinct from Elodea canadensis Michx. 



