508 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 83 



taking the measurements for this study, the tail was straightened 

 by pressure. When the tail was much coiled and contracted it 

 was sometimes necessary to exert considerable force to straighten 

 and hold it in place. No attempt was made to straighten the trunk. 

 The head was bent either upward or downward, as necessary, until 

 its lower profile assumed a position perpendicular to the tail. With the 

 fish held in this position, the length is taken to be the distance from 

 the tip of the tail to the tip of the coronet; the tail is the distance 

 between its tip to the median, lateral point on the boundary between 

 the last trunk and the first tail segment, as defined above; wliile the 

 trunk is the distance from the latter point to the upper margin of the 

 gill opening. The depth is that of the deepest segment, usually the 

 one on which the origin of the dorsal is situated, or the one immedi- 

 ately before or behind, this being in front of the brood pouch, and 

 consequently the measurement of males having that structure much 

 developed is fairly comparable with all other specimens. The depth 

 is taken between the lowest points on the upper and the midventra,l 

 ridge, the spines not included. The head is measured from the tip 

 of the snout to the dorsal margin of the gill opening; the postorbital 

 from the latter point to the posterior margin of the orbit; the snout 

 from its anterior margin, on the midline, to the anterior margin of 

 the orbit; the orbit between opposite points on its bony margin v.hich 

 is usually indicated as a slight keel but often quite indistinct (this 

 measurement, therefore, is not susceptible of a high degree of accu- 

 racy). Since the fish is more or less contractible or distensible along 

 its longitudinal axis, the length of the tail, the trunk, and the total 

 length used as a standard of comparison will vary somewhat wdth the 

 state of the specimen at the time of its preservation, and also with 

 the state of preservation. Tliis inaccuracy is inherent in the subject 

 matter and cannot be avoided. However, if the same method is 

 followed throughout, the figures are comparable, especially if they 

 are based on numbers of specimens. All measurements are expressed 

 as a percentage of the length. 



The methods of counting the segments and taking the measure- 

 ments are explained in detail, because it is absolutely essential to 

 follow a uniform system; otherwise descriptions of seahorses are not 

 of much value in the identification and distinction of the species. For 

 instance, the trunk segments in a given species may be recorded 

 either as 11 or 12, depending on the method of counting, as noted. 

 Since the predominant difference of one segment is usually a good 

 specific character, the necessity for care and uniformit}^ of method 

 is evident. All measurements recorded here were made Vvdth a 

 vernier caliper. 



