Extreme vertebral variation induced by temperature in a homozygous clone of the self-fertilizing cyprinodontid fish Rivulus marmoratus

Embryos from one clone of Rivulus marmoratus were reared at various constant temperatures ranging from 19.5 to 31.2C, sustained throughout early development, or else were transferred from 26 to 20C after various periods of development. Resultant vertebral counts were progressively lower at higher sustained temperatures (ranging from 34.94 to 32.57). Vertebral differences expressed as a percentage of the total count were greater in this (7.0%) than in previously reported comparable experiments on 14 other fish species (4.1% or less). In most previous experiments meristic variation may have been due, at least in part, to selective mortality acting on a varied gene pool. In R. marmoratus the high degree of genetic uniformity eliminates the likelihood of selective mortality as a cause of vertebral variation, which must therefore have been environmentally induced. Pectoral ray counts also were lower at higher sustained temperatures; other meristic series did not show sharp responses. Temperature transfer experiments showed that vertebral counts are determined within 4 days at 26C (by first appearance of retinal pigmentation). Pectoral ray counts are not fully determined until about 8 days, shortly before hatching. Temperature breaks produced "shock effect" in pectoral rays but not in vertebrae. The bearing of homozygosity on amplitude of temperature-induced vertebral variation is discussed.