142 Trans. Acad. Sci. of St. Louis. 



important characters gone there remains little to distinguish 

 the spurious specimen in hand from the narrow-leaved forms 

 of cordata. Even the veiuing, perhaps the most constant 

 single character of cordata, fails us here. Inasmuch as S. 

 sericea is widely spread over the country, is not its influence 

 the same wherever comin": in contact with cordata? And 

 may not many of our narrow, long-leaved forms be due to this^ 

 influence? 



A detailed technical description is not one of the objects 

 of this paper. I will call attention, however, to several 

 variations which seem, hitherto, to have been overlooked, or 

 which may be peculiar to our region. One is a tendency, in 

 some specimens, to the notching of the stigma, a good example 

 being represented in Figs. 19, 20, 21. Another is the occa- 

 sional tendency to union partways, of the filaments, repre- 

 sented in Fig. 18; both are from trees. Summing up the 

 three classes into which the collection is divided, it will be 

 seen that the oblong and long lance shapes, the obtuse to 

 truncate base, the medium petiole, the obtuse stipule, and 

 glaucous surface, predominate. 



In conclusion, allow me to say, it gives me no pleasure to 

 run counter to the results arrived at by Mr. Bebb. Had he the 

 same material to work with, I feel confident there would be no 

 essential difference. He wrote me he had barely a dozen speci- 

 mens amongst which were no truncate or cordate leaves, which, 

 of course, was very exceptional to all his experience of the 

 cordatee. This deficiency must be attributed to the fault of 

 the collectors ; for in the so-called Missouriensis, the collector 

 must look for the most vigorous shoots to obtain the truncate 

 and cordate leaves, inasmuch as the tendency in this direction 

 is not strong. It is not to be expected this paper will settle 

 the question at issue. It should provoke further inquiry. 



St. Louis, Nov. 4, 1895. 



