Genetisk studie av inavel och fertilitet hos hund
Inbreeding, the mating of related individuals, has been the one dominating breeding strategy for dog breeders during the last 200 years. However, its negative effects on fer-tility and other fitness traits, a phenomenon called inbreeding depression, have been ignored in the quest of perfect unity in conformation or performance. The purpose of this study was to test the hypotheses that litter size is reduced as a consequence of in-breeding, affected by both the inbreeding coefficient of the dam and the litter and that a rapid increase in the inbreeding coefficient affects litter size more than a slower rate.
Data were collected from the Swedish Kennel Club registry of origin. Breeds included in the study were the Bernese Mountain Dog, the Cocker Spaniel, the Golden Retriever, the Rottweiler and the West Highland White Terrier. Based on the information obtained from these origin files inbreeding coefficients, and the fraction of known ancestors per generation were calculated for all dogs using the Fortran Pedigree Program PEDIG. SAS and Microsoft Excel were used for all other analyses.
To get an overview of the whole material, diagrams of the distribution of a Pedigree Completeness Index, inbreeding trend over time and the distribution of close matings were all drawn. The results from these initial diagrams suggested that the research data be limited to circumstances where both litter and dam have a Swedish registration number; have at least 70% known ancestry in eight generations; and that the litter should be born between 1988 and 2002. From these limited data the distribution of litter size and the effect that different types of inbreeding had on litter size were investigated. The types of inbreeding were total; new (based on a three generation pedigree); and old (based on generations four and backwards). It was not possible to do multiple regres-sion analyses and therefore the effect that types of inbreeding had was estimated using simple regressions.
The results showed that a large part of the population within each breed had a high Pedigree Completeness Index on eight generations, although none had 100%. Generally the mean total inbreeding coefficient decreased with time. At least two thirds of all lit-ters in the material had parents that did not share an ancestor for three generations and only in the Golden Retriever there was an obvious increase in those matings. The dis-tributions of litter size usually displayed a bell-shaped normal distribution curve. The Cocker Spaniel had the highest observed inbreeding coefficient and the Bernese Moun-tain Dog had the lowest. Cross tables revealed that the most frequent combination in the material is the one where both litter and dam have a total inbreeding coefficient of 0 ? 0,05. Investigation of new inbreeding showed that except for the odd case inbred dams do not give birth to litters with a high inbreeding coefficient and that litters with a high inbreeding coefficient have a non-inbred dam. As for total and old inbreeding most ob-servations were found where the inbreeding coefficients of the litter and dam were ap-proximately the same, between 0 and 0,10.
The results did not generally support the hypothesis that a rapid rate of inbreeding is more serious than a slower one. However, the regression analysis showed that a high inbreeding coefficient, be it new, old or total, in both dam and litter resulted in a de-creased litter size. The effect was most significant for the inbreeding coefficient of the litter.