678 



STERNA FOESTERI, FORSTER S TERN. 



Differential diagnoses of S. forsteri, hirundo, and macfrura. 



S. forsteri : Bill (average), 1.60 along culinen ; depth at base, 0.40 ; robust. Bill 

 orauge-yellow, nearly its terminal half black. White space between eye and cutting 

 edge of upper mandible broad. Under parts white. Outer web of first primary sil- 

 A'ery ; the iuaer webs also of the others strongly Loary, without well-defined white 

 spaces. Tail bluish-pearl, like the back, its lateral feather greatly produced (average 

 nearly 7 inches in length) ; its outer web white, inner the color of the rest of the tail. 

 Legs long and stout. Length of tarsus (average) rather over 0.90 of an inch ; orange- 

 yellow, tinged with vermilion. Length of tarsus, middle toe and claw, 2 inches. 



S. hirundo : Bill (average), 1.45 along culmen ; depth at base, 0.33 ; moderate. Bill 

 vermilion-red, its terminal third black. White space between eye and cutting edge of 

 u)>per mandible narrower than in forsteri. Under parts lightly washed with plumbe- 

 ous, fading into white on the throat and abdomen. Outer web of first primary black ; 

 inner webs of the others somewhat hoary, with well-defined white spaces. Tail white, 

 dili'erent from the back, its lateral feather moderately produced (average inches in 

 length); its outer web grayish dusky, inner white. Legs moderate. Length of tarsus 

 about 0.80 of an inch ; light vermilion-red. Length of tarsus, middle toe and claw, 

 1.75 inches. 



*S'. macrura : Bill (average), 1.30 along culmen; depth at base, 0.30; slender. Bill 

 wholly deep carmine-red. White space between eye and cutting edge of upper mandi- 

 ble narrower than in hirundo. Under parts decidedly plumbeous, extending from vent 

 to throat, both of which become abruptly white. Primaries as in hirundo. Tail with 

 the elongation oi forsteri, or rather exceeding it (average 7.50 inches), and the color of 

 hirundo. Legs very short and sleuder. Length of tarsus (average) 0.65 of an inch; 

 deep vermilion, almost lake. Length of tarsus, middle toe and claw, about 1.50 inches. 



Comparison of the young-of-the-year of S. forsteri and hirundo. — The bill and feet con- 

 stantly present differences i>roportioual to those which exist in the adults, as regards 

 length and stoutness. The bill of hirundo is more decidedly yellowish at the base of 

 the lower mandible than in that of forsteri, and the feet are clear yellow instead of 

 being tinged with dusky. The mottled and variegated crown and upper parts are much 

 the same in both, and the markings of the quills quite identical. The tail, however, 

 ditfers remarkably. In hirundo the outer webs of all the feathers are diisky gray. In 

 forsteri the reverse is the case. The difference is even more marked than in the adults. 



Synonymy. — This species was first indicated by Swainson and Richardson, as above, 

 under the name of Sterna hirundo, these authors, however, apiireciating and com- 

 menting upon the differences from that species. Nuttall was the first to give it a spe- 

 cific name, based upon the indications attbrded by the authors aliove mentioned. To 

 Mr. G. N. Lawrence, however, is really due the credit of first establishing the species 

 by giving complete descriptions, and of showing its relationships to S. hirundo. This 

 misapplication of Linnieus' name by Swainson and Richardson is, I believe, the only 

 specific synonym (excepting Audubon's ^^havelli"), unless the Sterna nuttalli of Nnttall's 

 work (from Audubon's Mss.) is to be referred here, as is probably the case. In Bona- 

 parte's Table of Longipennes (C. R. xlii, 1856, p. 768) not only is havelli separated sjie- 

 cifically t'vonxforsleri, but it is placed in a different genus (Gelochelidon), while trudeaui 

 is made a Thalasseus. It is difficult to account for such a misapprehension of th( ir 

 affinities, but vagaries of this sort are too often found in the latest works of this illus- 

 trious author. It is unnecessary to reproduce here the argument by which I showed 

 (Pr. A. N. S. Phil. 1862, p. 543) the identity of havelli and forsteri, for the fact is fully 

 established. 



I append the measurements of several specimens of /or«/cn, showing the limits of 

 its variation in size : 



.5 3 



24274 



12692 



11624 



4928 



9973 

 13473 



43K 



Locality. 



New Jersev 

 ....do..-..\ 

 ....do 



Florida 



..-.do 



Sac Valley 



Utah 



California . 

 Louisiana. 



