THE GREATEST i^LYINO CREATTTRE. r>51 



the Pterodactyl. 'Pliis may have been (luite i!(> feet t'l'om ti]) to tip of 

 wing'. The pahv)ntoh)iiist says that approximately the wiiii;- surface 

 was 25 square feet, the weio-ht somethino- like ;>(» pounds, and I infer 

 from the consideration just ([uoted that the power was probably less 

 than O.Of) hors(>])ower; the immens(dy <>-r«^ater c^conomy and efKciency 

 of nature in the respi^qt of pow(>r })eino' most strikinuly shown by the 

 size of the small rectangle as comj)ared with that in the tlyino- machine 

 of man's invention. 



After this comes the condor, preeminently a soarer. Its stretch of 

 wing is 9 to 10 feet, its supporting" area very nearly 10 square feet, 

 its weight IT pounds, and the approximate horsepower it develops 

 (inferred from the facts already stated) scarcely 0.0,5. 



Next comes the turkey buzzard, whose stretch of wing is 6 feet, 

 its supporting area a little over 5 square feet, its weight 5 pounds, 

 and the approximate horsepower it develops (as al)()ve) 0.015. 



All the above are soaring birds. 1 now pass to another order of 

 birds, which Hap their wings. The wild goose, with a supporting 

 area of 2.7 s(juare feet, has a weight of 9 pounds, and needs a pi'opor- 

 tionately greater power of nearly O.020 horsepower to drive it, as 

 against scarcely 0.02 horsepower in the last examph>. 



Next we have another familiar bird, the pigeon, which drives itself 

 by flapping the wings. This has an area of about o.T of 1 square foot, 

 a w'eight of 1 pound, and a horsepower of 0.012. 



Below thi-i we come to the humming Ijird, whose area, l)eing shown 

 on the same scale as the others, is almost too small to be distinguished 

 on the page, but which has a supporting surface of nearly 0.08 of 

 a square foot, a weight less than 0.02 of a pound, and a horse- 

 power of probably not over 0.001. (All these values, as we have 

 already said, are but approximative.) 



Particular attention is to be paid to the fact that regarding the 

 ratios of supporting surface to weight supported, these ratios are not 

 only not the same in all the birds, l)ut themselves dift'er greatly, but 

 systematically, with the absolute weight. If we inquire how much 

 1 horsepower would support, for instance, supposing the ratios of 

 sustaining surface (i. e., wing area) to weight to l)e constant, we find 

 that 1 horsepower would, in the Hying machine, suppoi't 20 pounds 

 with 86 scjuare feet area of wing (i. e., If square feet to a pound); 

 and that, passing to the flapping birds, if the wild goose were to pre- 

 serve the same relations on an enlarged scale, its 1 horsepower would 

 support .346 pounds of weight with the use of 101 square feet of wing 

 surface or 0.29 square feet to the pound; that in the pigeon 1 horse- 

 power would support 83 pounds of weight with the use of 58 square 

 feet of wing surface or 0.7 square feet to the pound, and that in the 

 humming bird 1 horsepower would support 15 pounds of weight with 

 the use of 26 square feet of wing surface or 1.73 square feet to the 

 pound. So that, broadly speaking, so far as these few examples go, the 



