TREE PERNS OF NORTH AMERICA — MAXON. 489 



and size of leaves, fern students up to a comparatively recent time have 

 been dependent mainly upon the chance and often inexact observa- 

 tions of collectors, who as a rule have not been well acquainted with 

 the group, and upon the characters offered by conservatory specimens. 

 As has been pointed out, excellent distinguishing characters are 

 afforded by the trunk and by the vascular parts of the frond, especially 

 by the stipe bases, which are mostly armed with spines of different 

 lands for different species and clothed with equally characteristic 

 scales. Yet ] these are the very parts usually omitted from collections, 

 which are likely to consist only of a small section of the blade, often a 

 single pinna. To know from such scant material the real characters 

 of plants with stems 10 to 50 feet high and leaves 5 to 15 feet long is 

 obviously impossible; and many have been the complaints of fern 

 writers from Sir William Hooker's time almost to the present upon the 

 inadequateness of herbarium specimens. Within the past two or 

 three decades, however, and with the development of more convenient 

 and ready means of travel, the practice of sending investigators to col- 

 lect material of their own special groups has become general. Very 

 full and valuable data are being obtained in this way, in marked con- 

 trast to the incomplete and piecemeal information conveyed under the 

 old method; and it has been found feasible to collect and dry charac- 

 teristic sections of the trunk of most species, in addition to the stipes 

 and the lower, middle and upper pinnae of the fronds. It is upon the 

 basis of carefully selected and complete herbarium material of this 

 sort, supplemented by photographs and by exact field data as to size, 

 form, structure, and color of the various parts, that an adequate knowl- 

 edge of tree ferns will finally be built up. 



EXPLANATION OF PLATES. 



Plate 1. 



Group of tree ferns (Cyathea princeps), near Sepacuit6, Alta Verapaz, Guatemala. 

 Photographed by Mr. G. N. Collins, April 16, 1904. 



Plate 2. 



View from the Adjuntas Road, Porto Rico, showing plants of Cyathea arborea. Photo- 

 graphed by Mr. G. N. Collins, July 11, 1901. 



Plate 3. 



Two forms of trunk of Cyathea arborea (see p. 472), collected in the valley of the Rio 

 Bayamita, southern slopes of the Sierra Maestra, eastern Cuba, altitude 

 2,000 to 3,500 feet, April, 1907; fig. a, Maxon 3909 ; fig. 6, Maxon 3906. Both 

 one-half natural size. 



Plate 4. 



Coffee warehouse at Sepacuit6, Guatemala; excepting the five heavy supporting tim- 

 bers the upright sides are entirely of tree fern trunks. Photographed by 

 Mr. G. N. Collins. 



