228 " GENERAL OBNITHOLOGY.. 



will be found, generally down to species and even varieties. They are to be used as follows 

 (after the preceding lessons have been learned) : — 



We have in hand a bird we do not know, and the name of which we wish to ascertain. 

 Suppose it to be that common species which buUds the nest of mud upon the bough of the 

 apple-tree and lays greenish-blue eggs. To what family does it belong? 



The Key opens with an arbitrary division of our birds according to the number and 

 position of their toes. Our specimen, we see, has four toes, three in front, one behind. It 

 therefore comes under IV. Going to IV., we read : 



Hind toe — inserted above the level of the rest, etc. 



— not inserted above the level of the rest. . . . (GotoB.) 



Our specimen has the hind toe not inserted above the level of the rest. Going to B, we find 

 five alternatives. Our bird presents no one of the special characters of the first four alterna- 

 tives, and this determined takes us to g. There we find : 



(g) Primaries — 10 ; the 1st (never spurious), etc. 



— 10; the 1st (spurious or), etc. . . . (Gotoi) 

 — 9 ; the Ist (never spurious), etc. 



In this case the bird has obviously a spurious first primary, not nearly two-thirds as long as 

 the longest. Going to * ; — 



(i) Tarsus— " booted " ; wings— shorter than, etc. 



— longer than tail ; tail — double rounded. 



— not double rounded TnKDiD.a;, p. 240. 



Thus (provided we have taken the trouble to inform ourselves what " spurious first pri- 

 mary " and "booted tarsus" mean), the key conducts to a family, by presenting in succession 

 certain cdternaUves, on meeting with each of which, we have only to determine which one of 

 the two or more sets of characters agrees with those afibrded by our specimen. There will 

 not, it is believed, be any trouble in determining whether a given character is so, or is not so, 

 since only the most tangible, definite, and obvious features have been selected in framing the 

 key. After each determination, either the name of a family is encountered, or else a reference- 

 letter leads on to some new alternative, until by a* gradual process of elimination the proper 

 family is reached. After a few trials, with specimens representing different groups, the process 

 will be shortened, for the main divisions vrill have been learned; still the student must be 

 careful how he strikes in anywhere except at the beginning, for a false start wiU soon set him 

 hopelessly adrift. The key has been tested so thoroughly that there is little danger of his 

 running oflf the track except through carelessness, or misconception of technical terms; but 

 there is no excuse for the former, and the latter may be obviated by the Glossary at the end of 

 the book, and especially the foregoing General Ornithology, 5 3, which should be consulted 

 when any doubt arises. Time spent upon the preliminary lessons wiU be time saved in 

 the end. 



At page 240, as indicated, the family Turdidce is fully characterized, and its sub-families 

 and genera are analysed. The bird in hand should answer all the characters of the family and 

 those of one of the sub-families, Turdin<B, and one of the genera, Turdus. The analysis of 

 the species of Turdus should show the specimen to be Turdus migratorius, the Robin. Under 

 the head of that species. No. 1 of the List, wiU be found a fair description and various other 

 particulars. 



If there be any difficulty in going at once to the family, the student may try the key to 

 the orders and sub-orders, and get on the track in that way. 



Directions for measurement have already been given (p. 24). In comparing measure- 

 ments made with those given in the Synopsis, absolute agreement must not be expected ; 

 iodividual specimens vary too much for this. It will generally be satisfactory, if the discre- 



