SECRETARY'S PORTFOLIO. 391 



the white of the corolla. As a merely oriKiinental plant the Speciosa is 

 superior in bloom as well as in habit of the tree. 



Fruit — Tlie fruit-})ods, often called Indian beans, vary in size and length 

 on trees of either kind, but those of the western tree may attain tiie extreme 

 length of twenty-four inches, and they are of a different shape when examined 

 on the cross-section. This is cylindrical, while that of the Geoi'gia kind is 

 elliptic; the valves meeting form an angle, and their margins project so as to 

 be pei'ceptibly felt as a ridge when the fruit is rolled or drawn between the 

 thumb and fingers. The pods of the Speciosa are usually of a dark brown 

 color, and the valves are marked externally by parallel grooves that extend to 

 their entire length. Those of the species, as in cultivation, are usually of a 

 lighter color (rusty brown), and less distinctly grooved, in some scarely in- 

 dented by shorter channels. 



Seeds — The distinctive characters derived from these curiously winsred 

 organs are the most reliable and constant, especially the arrangement of the 

 margins and terminations of the wing or projecting membrane. In size the 

 seeds of the Speciosa exceed both in length, breadth and weight.* The coma 

 or tuft of hairs at each end of the seed of the species is compressed and pointed 

 as though it had been wetted and drawn together, while in the A\'estern kind 

 they are separate, parallel and almost digitate. 



The color of the membrane is also lighter in the Eastern kind, and the tex- 

 ture firmer and more satiny, while that of the Western kind is softer and more 

 silky. Tliese distinctions were first pointed out by the practical seed-handler, 

 Mr. Robert Douglas, of the Waukegan nurseries, in Illinois, whose acumen as 

 a judge of tree seeds is remarkable, the result of long experience and close 

 scrutiny, particularly of the seeds of conifei'se. 



Hardiness of the tree — This character is of the greatest importance to the 

 practical and extensive tree planter, for though we may be willing to put np 

 with a so-called half-hardy tree upon the lawn or in the ornamental planta- 

 tion, on account of its beauty or rarity, the forester cannot afford to run any 

 risks, and should only select such trees for his economical groves as have an 

 unquestionable record in this particular. 



Range — An approximate limit to the northerly range of these plants is all 

 that can now be attempted, and many exceptional cases may be cited where 

 individuals have escaped injury from frost even beyond the limits that will be 

 indicated. 



In Eastern Massachusetts there are old trees of the common kind that have 

 stood for seventy-five, or, perhaps, one hundred years; and yet Professor C. 

 S. Sargent, who kindly reported them, says that he '* cannot recommend it as 

 a forest tree for that region, except in sheltered situations, because it does not 

 always perfectly ripen its fruit," which he very probably considers a safe test; 

 and yet it is one that would preclude the planting of alpine hei^jhts in Europe 

 that are now clothed with valuable timber. 



The western kind has a few representatives about Falmouth, in Barnstable 

 county, in the same State, as reported by the successful tree-plantei*, Jos. S. 

 Fay, of Wood's Holl, Mass. 



Westward from ihe Alleghanies the safe limit comes down to a lower range, 

 as will be indicated by a few citations. At Toledo, Ohio, latitude 41, it is cut 

 to the gi'onnd, and even at Dayton young trees of the common kind often suf- 

 fer from the cold. In Michigan they generally suffer ; at Fort Wayne, Ind., 



*26 seeds Speciosa (from Vincennes) = 1 1 jris. 

 25 seeds Georgia (from Terre Haute) = i) grs. 



