MR. E. DEGEX 01^ ECDTSIS. 353 



also proved to be of no inconsiderable value as a foundation in the compilation of the 

 comparative diagrams for the " Remiges " (see Pis. XXXVI. -XXXVIII.). 



These measurements, moreover, have been productive of a number of useful obser- 

 vations, which might otherwise have escaped notice. I may mention, for instance, the 

 case of the almost isolated condition of the development in rapid succession of primary 

 remiges III., IV., and V., which attain their full lengths without having a correspond- 

 ing equivalent on the cubital portion of the wing (see Chart, Table II., p. 370, and 

 Diagram B, Plate XXXVI. , left and right wing, showing the relative growth of the 

 flight-feathers, which are lettered <?, d, and e respectively). 



This method has further afforded the means for estimating in percentages the 

 reduction in the power of flight through the loss of flight-feathers during the various 

 stages in the moult of the bird (see Table III., p. 374). That the length of corresponding 

 remiges is never alike in any two individuals seems conclusive, and it is enough, for 

 all practical purposes, to ascertain their mean length. These averages were for the 

 Metacarpo-digitals or Primaries : — 



I. II. HI. IV. V. VI. VII. VIII. IX. X. 



mm 135 142 149 159 182 195 205 197 178 96 



for the Cuhitcds or Secondaries : — 



I. 11. III. IV. V. VI. VII. VIII. IX. X. 



mm 132 127 124 120 115 110 100 95 75 55 



Possibly this procedure may appear somewhat artificial, but it has the advantage of 

 at least supplying approximate data for the ultimate length of those remiges which, 

 being in a state of partial development (or frequently shed altogether previous to their 

 replacement), in this way can be made available for exercising a control over the 

 deficiency caused by their loss. 



In this examination some standard is necessary ; therefore, each flight-feather has been 

 theoretically reduced to the uniform length of 100, which figure represents its full length. 



How this works out may be seen by a reference to the diagrammatic representation 

 of specimen No. 1 in Plate XXXVI. (left wing), and further marked A, in the 

 case of primary I., whose actual length is 110 mm. (expressed in the space on the 

 top), whilst its length in proportion to that of the full growth (i. e. 135 mm.) (see above) 



is therefore — j^— =81-0, or in round figures 80, which is practically f of its 

 ultimate length. Or to take the case of remex X., which is the shortest of the 

 primaries, whose mean length was ascertained to be 96 mm., as shown in specimen 

 No. 25, and further marked K on Plate XXXVIII. (see Table as in former), which repre- 

 sents this particular stage. Since its actual length was 65 mm. and its mean computed 

 at 96 mm., the relative length therefore is, according to the adopted rule, '^ ,^^ = 677, 

 or practically 70 — that is to say, -^^ of its ultimate length. 



