Eggs were stripped from gravid Atlantic silversides collected on two occasions. once during the early part and once during the late part of the natural spawning season. Unfertilized egg diameter was not correlated with length of the female. nor was it significantly larger during the early part of the season. Eggs were fertilized and incubated in the laboratory. Larval length at hatch was measured every 24 h during the hatching period after embryos were incubated at I8 or 25" C. Lower incubation temperature caused a significantly greater length at hatch for the offspring of each ofthe 20 females studies. In most cases (17 out of20 at 25" C, 10 out of 20 at 18" C). there was a significant decrease in length at hatch during the hatching period for a given female's eggs incubated at a given temperature. In the natural environment. larvae hatched early in the season under cooler temperatures could average 12% longer than those hatched later under warmer tempcratures, and therefore may have a greaterchance ofsurvival. The results help to explain the observation that field-caught M . nienidia that hatched early in the season are larger at any given age than those that hatched late in the season.

Otoliths were removed from fieldcollected silversides of age less than 3 months. Otolith diameter was highly correlated with total length of the fish. Daily growth ring counts for this species are known to be a function of age rather than size, so widths for the daily growth rings provide a record of daily increases in length of the fish. Measurement of ring widths showed that weekly specific growth rate was greater than 70% at age 1 week, but declined to about 30% at age 1 month and about 15% at age 2 months. A laboratory experiment in which temperature was changed on a weekly basis demonstrated that environmental variables can affect the width of rings. Nevertheless, the growth rate of field-collected fish, as calculated from otolith ring widths, was more highly correlated with size of fish. as measured by otolith radius, than with the environmental variables of temperature, salinity and plankton abundance. Back-calculation of growth rates from otolith ring widths of five fish collected at the end of the growing season yielded the same age-growth curves as were obtained from 203 fish collected biweekly during the season.

No abstract

No abstract