-6 4 - 



dark green, lanceolate or narrowly triangular, bipinnate, or towards 

 the base tri-pinnatifid, many times forked at the apex, forming a 

 tassel about one inch long; pinnae narrow-lanceolate, one to 

 many times forked at the tip, tassel under half an inch long; pin- 

 nules oblong, toothed or divided nearly to the rachis ; under sur- 

 face and rachis markedly glandular- pilose; sori as in type. Type 

 specimen No. 253,636 in the U. S. National Herbarium, which is 

 one of a limited number distributed by Mr. Floyd as No 199. 

 Collected at Wildcat Notch, Blue Hill, Milton, Mass., Sept. 11, 

 1898. 



Concerning the occurrence of this fern Mr. Floyd writes: "I 

 found a large clump of them on the west slope of Blue Hill. It 

 occupied a corner of a much larger patch of the same fern that 

 was merely normal, and it attracted my notice immediately as I 

 was wading through the main clump. There were dozens of the 

 plants thus forked and I should say that they were spreading by 

 means of their creeping rootstocks. A very noticeable feature 

 was the unusual color of the crested form. They could be told 

 from the normal plants at some distance by their darker green 

 foliage." Years ago Mr. Davenport collected a crested form of 

 punctilobula on the same slope of the same hill, and Mr. Floyd 

 believes the two stations to be very close together. If this be 

 true, cristata may be on a fair way to become a permanent addi- 

 tion to our flora. At present it must be regarded as an interest- 

 ing variation apparently stable in a restricted locality, and bid- 

 ding fair to spread successfully. 



In the foregoing I have followed Lawson in his use of the 

 generic name Dennstoedtia in the " Fern Flora of Canada." The 

 citation for the species should read DennstcEdtia punctilobula 

 (Michx. ) Moore. Dicksonia, Prof. Underwood writes me, is to 

 be applied wholly to the southern hemisphere tree ferns ; and the 

 genus Silolobium, established by Desvaux upon our species as a 

 type, can not be regarded as generically distinct from the tropi- 

 cal forms with creeping stems which come under the genus 

 Dennstcedtia, established by Bernhardi. It becomes necessary 

 then to follow Moore and Lawson in adopting Dennstccdtia as a 

 generic name to replace Dicksonia as applied to our North Amer- 

 ican D. punctilobula. 



Washington, D. C. 



