MAY: EFFECTS OF DELAYED INITIAL FEEDING 



convenient material for the study of embryonic 

 and larval development. 



Eggs for the present study were collected dur- 

 ing a spawning run on the night of March 24, 

 1970, at the beach in front of Scripps Institution 

 of Oceanography, La Jolla, Calif. The eggs from 

 a running ripe female were expressed into a 

 small plastic container and artificially fertilized 

 by adding milt from one male and a small amount 

 of seawater; after approximately 1 min the 

 water was decanted and sperm removed with 

 several washes of fresh seawater. At the Na- 

 tional Marine Fisheries Service Fishery-Ocean- 

 ography Center in La Jolla, developing eggs 

 were dispersed in a layer of washed, slightly 

 moist beach sand approximately 1 cm deep at 

 the bottom of rectangular plastic containers 

 (16 X 12 X 11 cm), and a paper towel moist- 

 ened with seawater was placed on the surface 

 of the sand. The tops of the containers were 

 covered with aluminum foil. This incubation 

 procedure, essentially the same as one described 

 by Morris (1956), kept the eggs moist and pro- 

 duced good hatching when excess water (which 

 quickly brought on anoxic conditions) was 

 avoided. The containers were placed in a water 

 bath held at 20° ± 1° C by manually mixing 

 water from the warm and cold seawater systems 

 of the Fishery-Oceanography Center (see Las- 

 ker and Vlymen. 1969). The day before hatch- 

 ing, the temperature of the water bath was 

 lowered to 18° C over a period of 3 hr. 



On April 3, 1970, after 10 days of incubation 

 (a common incubation time in nature, according 

 to Walker, 19.52), hatching was induced by ad- 

 ding filtered seawater at 18° C to the incubation 

 containers and agitating the water and sand by 

 drawing water i-apidly in and out of a pipette. 

 Hatched larvae were immediately transferred 

 via pipette (4-mm bore) to 18° C water in rear- 

 ing containers. In this paper the day of hatch- 

 ing will be referred to as day 0, the day after 

 as day 1, and so forth. 



DESIGN OF EXPERIMENT 



A total of 20 rearing containers was set up, 

 each with approximately 50 newly hatched grun- 



ion larvae. Seven containers (*l-7) were used 

 to determine the efl"ect of delayed initial feeding 

 on mortality and growth; in six containers in 

 this group, feeding was begun at progressively 

 later times at 3-day intervals beginning on day 

 1— i.e., on days 1, 4, 7, 10. 13, and 16— while the 

 larvae in one container (-»7) were not fed and 

 served as a control.. On the 20th day after 

 hatching, all surviving larvae which had received 

 food in this series were collected, and the length, 

 weight, and chemical composition of the larvae 

 were determined. Seven of the other containers 

 (#8-14) were fed daily beginning on day 1, and 

 six (*15-20) were given no food; these con- 

 tainers, referred to as "supply containers" in 

 what follows, supplied fed and unfed larvae for 

 measurements of length, weight, and chemical 

 composition and also for experiments on feeding 

 and growth. On the same days when feeding 

 was begun in a new container in the delayed 

 feeding series (containers *l-7), larvae from 

 both the "fed" and the "unfed" supply containers 

 were used to determine the incidence of feeding 

 and to begin quantitative feeding experiments. 



PHYSICAL CONDITIONS 



Water of approximately 33 '/,, salinity was 

 taken from the seawater system of the Fishery- 

 Oceanography Center. The larval fish contain- 

 ers were originally filled with HA Millipore- 

 filtered seawater, and at weekly intervals 

 filtei-ed water was added to the fed containers in 

 order to replace water removed with uneaten 

 food and fecal matter (see below), the volume 

 added usually being between 1 and 2 liters. The 

 temperature in the water bath was kept at 18° 

 ± 1° C, and temperatures in the larval contain- 

 ers were within 0.5° C of the bath temperature. 

 Eighteen degrees is the midpoint of the 14°-22° 

 C range of water temperatures which occurs off 

 La Jolla during the spawning season of the 

 grunion (Reid et al., 1958). Banks of two 40- 

 watt "daylight" fluorescent lamps ]30sitioned 76 

 cm above the surface of the water illuminated 

 the containers for 12 hr each day. The lights 

 were timed to go on after sunrise, so that dif- 

 fuse light entering through windows increased 



413 



