-58- 



base, the keel being arched. They also differ in the fact that the dorsal lobe, 

 which covers over the basal pocket, is shorter than the others and suberect. 



The production of gemmae is an exceedingly common phenomenon in S. 

 scitulus. They are borne on the apices of leaf-lobes, and sometimes a long 

 succession of gemmiparous leaves may be observed. Eventually, however, 

 the growth of the shoot is brought to an end and the last leaves produced re- 

 main rudimentary. Apparently the leaf-lobes are normally acute, and the blunt 

 lobes which are so often present indicate that the leaves have been gemmiparous. 

 The gemmae themselves are golden-brown, angular bodies, with slightly thick- 

 ened walls (fig. 5). They are usually divided into two cells and measure 20- 

 25 M in length. 



The gemmae and leaf-cells of S. scitulus are very much like those of 5. ex- 

 sectceformis , although the trigones in the second species are not always well 

 developed. In 5. exsectceformis, however, there is a sharp disparity in size be- 

 tween the tooth-like dorsal lobe and the ventral lobe, while the latter, although 

 sometimes shortly bilobed at the apex, is often acuminate and undivided. 

 Explanate leaves bring out the differences between the lobes even more strik- 

 ingly than leaves in their normal position on the stem and are in marked con- 

 trast to the leaves of S. scitulus with their three subequal lobes. Another close 

 relative of 5. scitulus is 6". politus (Nees) Steph., which also is characterized by 

 equally three-lobed leaves and by leaf-cells with conspicuous trigones. But 

 the leaves in 5. politus tend to be scattered along the stem, the lobes are blunter 

 than in S. scitulus, and the leaf-cells are larger, measuring (according to Miiller) 

 25-45 M in the middle of the leaf and still more at the base. S. politus 

 is widely distributed in Europe and is also cited in the literature from Greenland, 

 Alaska, Ellesmere Land, Quebec, and British Columbia. 



4. Cephalozia Francisci (Hook.) Dumort. 



Collected in May, 1910, near Yarmouth, Nova Scotia, on earth along a 

 roadside, by J. Macoun {No. 41). The second record for North America. The 

 jfirst record was based on specimens collected by Mrs. A. R. Northrup in the vi- 

 cinity of Prospect Harbor, Maine, in August, 1902. These specimens were 

 reported upon by Miss Haynes.^ A third station for the species, also in Maine, 

 was discovered by the writer at Eastport, in August, 191 1. In northern Europe 

 C. Francisci is widely distributed, although not abundant. 



5. ScAPANiA PORTORicENSis Hampe & Gottsche. 



Collected in April, 1902, on Turquino Mountain, Santiago de Cuba, by S. 

 H. Hamilton {No. 5875). The first record for the island of Cuba. The species 

 was originally described from specimens collected by Schwanecke in Porto Rico 

 but is now known also from Jamaica, Colombia, Ecuador, Bolivia, and Brazil. 

 In Miiller's Monograph of the genus Scapania the plant is described and fig- 

 ured in detail with a full synonymy. 2 



1 Torrej'a 3: 40. 1903. 



2 Nova Acta Acad. Caes. Leop. Carol. 83: 150. pi. 18. 1905. 



