are valid indicators of age. (Later sections in- 

 clude considerable supporting data on certain of 

 the points.) 



1. Gizzard shad known to be young of the year 

 had no annuli on their scales. 



2. The scales of these young shad captured in 

 the fall, winter, and early spring showed no an- 

 nulus at the edge, whereas those taken later 

 exhibited increasing percentages with an annulus 

 until all possessed them. 



3. The distance between the last annulus and 

 the scale edge increased through the growing 

 season. 



4. Presumed age groups as indicated by modes 

 in length-frequency groupings of shad agreed with 

 groupings based on the number of annuli. 



5. The calculated lengths for particular years 

 of life among the age groups of the same, and 

 different years of collection agreed with estimates 

 of length from modes of frequency distribution. 



Body- Scale Relation 



The relation between fish length and scale length 

 in gizzard shad was determined from the "key- 

 area" scales of some 700 fish, ranging in standard 

 length from 43 to 390 mm. A test plotting of the 

 length of the anterior radius (center of focus to 

 anterior edge) of the scale against the standard 

 length of the fish indicated the relation to be 

 linear. The regression line, Z=22.1-f 44.25 S, 

 was fitted by least squares; Z=standard length 

 of the fish in mm. and S= anterior radius of the 

 scale in mm. Fish captured in every month and 

 including age-groups I through VI of both sexes 

 are represented in this equation. Studies of the 

 body-scale relation for each sex, for each age 

 group, and during each month revealed no appre- 

 ciable variation from the general equation. 



The empirically derived body-scale equation 

 was the basis for construction of a nomograph for 

 the calculation of growth from scale measurements. 



AGE COMPOSITION 



Seasonal movements, segregation by size and 

 possibly by maturity, and selectivity of collecting 

 methods complicated the problem of sampling 

 for age composition. The very young gizzard 

 shad of western Lake Erie were found in mid- 

 summer close to shore, usually in shallow water. 

 Collecting representative samples of these fish 

 presented difficulties. Capture of the very 



youngest required the use of dip nets which, of 

 course, the larger fish eluded. As the season 

 progressed and the fish grew larger, Common 

 Sense seines and bag seines were used, but this 

 gear permitted the very smallest fish to pass 

 through the meshes while many of the larger 

 ones escaped capture by their agility and speed. 

 The use of rotenone was selective for the very young 

 shad that could not swim through the poisoned 

 water rapidly enough to avoid being overcome. 

 (Rotenone was used in small areas in the lake 

 with no provisions for holding the fish within the 

 sampling area or for preventing dissipation of the 

 poison into the surrounding waters.) Electrocut- 

 ing, on the other hand, was selective for the larger 

 shad — the very young ones were unaffected. 

 Finally, I used dynamite. This method proved 

 must successful because fish of all sizes and ages 

 surfaced within the radius of its effectiveness. 



As the young gizzard shad become larger 

 (and older) they move into deeper water offshore. 

 By October, the Common Sense seine, which 

 must be fished inshore, will capture some O-group 

 shad but no I-group or older ones; hence, its 

 catches yielded no data on age composition. 

 The gill net captures large O-group shad and also 

 all the older age groups in October, but many 

 smaller O-group shad must pass through the 

 net. Because of these sampling problems, esti- 

 mates of age composition are unreliable among 

 the 4- to 8-inch shad. Only a few fish of these 

 sizes were captured by experimental gill nets 

 whose meshes ranged from 1 to 4 inches, stretched 

 measure. 



Except during the spawning season, gizzard 

 shad older than age-group III were rarely found 

 in the shallower waters. This preference for the 

 deeper water seemed to increase with age. 



In view of the segregation by size and the sea- 

 sonal changes of distribution, small-scale methods 

 of collection could not provide suitably accurate 

 data on age composition. I lacked especially 

 the means to compare data on the 1- to 4-inch 

 group from the shallow waters with records 

 for larger shad from the deeper waters. 



Seasonal Changes of Age Composition 



Because of the difficulties just outlined, O-group 

 gizzard shad caught by any means other than by 

 trap nets and 4-inch-mesh gill nets have been 

 excluded from this section. The O-group shad 



:;<tx 



U.S. FISH AND WILDLIFE SERVICE 



