xvi PLAN OF THE WORK 



The maker of keys, however, should not try to serve two masters. 

 If the keys will identify, they will have accomplished their purpose. 

 The classification of our birds is shown in the body of the book, where 

 the species are arranged according to the system adopted by the American 

 Ornithologists' Union. The keys proper are dichotomous, but 1 have 

 not hesitated to divide the larger families into three, or even four, 

 primary sections, the characters of which are placed together at the 

 head of the keys to permit of direct comparison. The heading of a 

 group or section applies to all the species included in it. Thus, if I 

 does not apply to the bird you are identifying, you must pass at once 

 to II, or III, or IV, as the case may be. Even should the first section 

 describe your bird, it is well to read the sections which are contrasted 

 with it. 



As a model to show how the keys are used, let us identify a male 

 English Sparrow, a bird which has a black patch on the throat and 

 breast, a slate-colored 'crown, chestnut marks on the sides of the head, 

 and a streaked back. Turning to the key to the family FringillidaB, 

 or Finches, Sparrows, etc., we read first the headings to the main 

 divisions, which are placed at the beginning of the key in order that 

 they may be more readily compared. 



"I. Underparts with red. 



"II. Underparts with no red, and without distinct streaks; throat 

 or breast sometimes with a patch or spot. 



"III. Underparts without red, and with numerous streaks." 



Our bird evidently belongs in Section II. Under this section we 

 read first: "1. Tail with white spots, bars or patches." Our bird has 

 no white in the tail, so we go to "2" and read: "Tail without large white 

 spots or patches." The first subsection here is: "A. Back plain, without 

 distinct streaks." Our bird is unquestionably streaked, so we therefore 

 go to "B. Back distinctly streaked." Under B we have "a. Bend of 

 the wing yellow," and "b. Bend of the wing not yellow." Our bird 

 has no yellow on the wing, and we therefore place it in subsection "6." 

 This subsection is subdivided into "6 1 . Crown bright reddish brown," 

 etc., and "c 1 . Crown not bright reddish brown." Evidently our bird 

 belongs in the group headed "c 1 ," which is further divided into "c 2 . Crown 

 streaked or spotted with black or black and white" and "d 2 . Crown 

 mixed grayish brown, and rufous, ashy or slate-color, without black 

 streaks." It is clear that our bird with its slate-colored crown, belongs 

 in the second of these sections. This, we observe, contains four species, 

 and we soon discover that "d 6 . Crown slate-color; a chestnut patch 

 behind the eye; throat black," fits our bird, and to prove the identifi- 

 cation we turn to the detailed description beyond. 



If this more than usually complicated case has been clearly demon- 

 strated, you should have no difficulty in using the keys, so far as 

 their construction goes. One has only to remember that if the descrip- 

 tive matter following I does not fit, one should try II; similarly 1 

 is contrasted with 2, A with B, a with 6, a 1 with 6 1 , 6 2 with c 2 , etc. 



