levels of both hybrids ranged from 60 to 70%. 

 These diminished levels in the hybrids may have 

 resulted from a combination of several factors 

 such as: the low male to female ratios used ( <1:1); 

 decreased sperm motility in the viscous hen-yolk 

 medium; high sperm mortality due to time, star- 

 vation, t^imperature shock, handling, etc. or par- 

 tial reproductive isolation between the species in 

 the form of mild fertilization block to non- 

 conspecific spermatozoa. 



The grunions, L. tenuis and L. sardina, showed 

 similar developmental rates (Moffatt 1977). De- 

 velopment proceeded normally in the hybrids and 

 at about the same rate as the controls. No unusual 

 embryonic mortality was observed in the hybrids, 

 evidence that these embryos were not gynogenetic 

 hybrids (Moore 1955). 



Preliminary trials showed that hen's yolk and 

 seawater alone will not initiate cleavage in Gulf 

 grunion eggs. Precautions were taken to prevent 

 conspecific milt contamination. Preliminary 

 examination of cellular nuclei smears of develop- 

 ing embryos immersed in colchichine revealed 

 somatic chromosome numbers of about 2n = 40 in 

 all four sets of embryos (controls and hybrids), 

 further evidence that these embryos were true 

 diploid hybrids. 



Grunion embryos will hatch after vigorous agi- 

 tation in seawater. On 31 March at 284 h (11.8 

 days) postfertilization, 65.6% of the L. sardina x 

 L. tenuis embryos hatched and 66.5% of the L. 

 tenuis x L. sardina embryos hatched at 272 h (1 1.4 

 days). These hatch times are similar to those of the 

 controls. Hatching can be induced in both grun- 

 ions at 10.2 days postfertilization when embryos 

 are incubated at 20° C (Moffatt 1977). 



Newly hatched L. tenuis larvae are typically 

 more darkly pigmented; they have a larger eye 

 diameter; they are stronger swimmers; and they 

 are more capable of escaping net capture than 

 newly hatched L. sardina larvae (Moffatt 1977). 

 Leuresthes tenuis larvae are 10% longer (mean 

 total length = 7.70 mm) than those of L. sardina 

 ( mean total length = 6.93 mm). The greater length 

 of the California grunion yolk-sac larvae occurs in 

 the postanal region as in the adults. California 

 grunion larvae are also 52%. heavier (mean dry 

 weight = 0.340 mg) whereas, the mean dry weight 

 ofL. sardina equals 0.223 mg (Moffatt 1977). The 

 greater length and weight of the California grun- 

 ion at hatching may be attributable to the 4.10 

 times greater ovum volume (Moffatt 1977; Moffatt 

 and Thomson in press). 



These differences which distinguish the prolar- 

 vae of L. tenuis and L. sardina were also observed 

 in the hybrids. In most characteristics the L. 

 tenuis x L. sardina larvae were not visibly distin- 

 guishable from the maternal controls (L. tenuis), 

 e.g., size, pigmentation, and swimming ability. 

 However, the L. sardina x L. tenuis larvae ap- 

 peared to be somewhat intermediate to the con- 

 trols in extent of pigmentation and swimming 

 ability. At 2 wk after hatching the length and 

 pigment differences between the larvae were more 

 pronounced. Premaxillary teeth were visible in 

 the L. sardina x L. tenuis larvae but not in the 

 reciprocal hybrids. Again, hybrids closely resem- 

 bled the maternal controls. Gulf grunion adults 

 typically have much stronger dentition than do 

 the adults of the California grunion (Moffatt and 

 Thomson 1975). 



As previously mentioned, the most diagnostic 

 differences between the adult grunions are the 

 lateral scale row counts. Scale counts of the 141- 

 day-old controls were essentially the same as 

 those of the adults (Table 1). The counts of the 

 hybrids were intermediate and significantly dif- 

 ferent from each other. Those shown by both hy- 

 brids were significantly different from those of both 

 parental species. The lateral scale rows of the hy- 

 brids were closer in number to those of the mater- 

 nal controls. Mean counts of L. sardina x L. tenuis 

 were 32% closer to those of L. sardina; and L. 

 tenuis x L. sardina were 20% closer to L. tenuis 

 than to those of the paternal parents, L. tenuis and 

 L. sardina. respectively.^ The intermediate counts 

 indicate paternal genome influence and that these 

 are indeed diploid hybrids. 



'A mean hybrid count greater or less than 65 (the midvalue 

 between the parental species) indicated the affinity to one parent 

 or the other. The numerical affinities (percentages) were calcu- 

 lated as the ratio of the differences between 65 and the hybrid 

 count and between 65 and the adult counts. 



Table l. — Means, ranges, n, and P values of lateral scale row 

 counts observed in 141-day-old hybrids and controls and adults 

 of the grunions, Leuresthes tenuis and L. sardina. 



'Student's f-test comparison of the lateral scale rovi( counts between the two 

 hybrids P- 001. 



478 



