This is an older paper but it seeks to explain the observation that some traits have more or less heritability than others and that survival traits may in fact have lower heritability than traits related to, for example, appearance.

Quoted text follows:

The hypothesis that traits closely associated with fitness will generally possess lower heritabilities than traits more loosely connected with fitness is tested using 1120 narrow sense heritability estimates for wild, outbred animal populations, collected from the published record. Our results indicate that life history traits generally possess lower heritabilities than morphological traits, and that the means, medians, and cumulative frequency distributions of behavioural and physiological traits are intermediate between life history and morphological traits.
 
There are many factors which may be responsible for the maintenance of genetic variance of fitness characters. The rate of origin of variation by mutation alone may be sufficient to maintain substantial additive genetic variance within natural populations. Heterozygote advantage, frequency dependent selection , variable selection in heterogeneous environments, diversifying selection, and migration have been proposed as possible mechanisms for the sustenance of genetic variation. Also, significant heritabilities for fitness characters are not inconsistent with zero additive genetic variance in fitness given negative genetic correlations between fitness components. Much support has been generated for this hypothesis; however others have challenged its generality. 

In conclusion, much of what has been generally accepted in the evolutionary literature concerning patterns of genetic variance in natural populations is verified by this study; fitness components do indeed generally possess lower heritabilities than traits believed to be unconnected to fitness. However, the mechanisms by which the high observed heritabilities of fitness components are maintained remains unclear. 

Mousseau, T. A., & Roff, D. A. (1987). Natural selection and the heritability of fitness components. Heredity, 59(Pt 2), 181-197.

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html