The Georgie Project has as its goal the analysis of genotypes and phenotypes with the intent of understanding the genetic basis of both normal and pathological (diseased) Portuguese Water Dog characteristics. To do this, we study the entire PWD population as well as its ancestors, using computer techniques specially developed for the project. This is very effective because the gene pool is small (few founders) and the population is large. The genetic database consists of pedigree information, which has been painstakingly gathered from existing records with the help of owners and breeders, as well as molecular data (DNA markers) which are currently being collected from living dogs. This last process is laborious and expensive, but already has demonstrated that the pedigree data is correct. Data has already been collected relating to the pathological conditions of GM1, PRA, Addison’s disease, Hip Dysplasia and now Cardiomyopathy.

This survey represents the first attempt to collect characteristics of normal dogs. We needed data on hundreds of dogs and received data on 360!!! These came from about 35% of the owners contacted and on the average those that responded sent data on two dogs (many on three or more).

The purpose of the survey was to obtain quantitative data on how much variation really exists in the PWD population. That information, in turn, would give us our first estimate of how different were the genes in the small gene pool. We realized that since all of you were making independent measurements there could be a lot of error in the measurements. However, from the pedigrees and your data we could calculate the heritability of the phenotype and thus know if your measurements gave us an indication of genetic differences (as opposed to measuring errors or differences resulting from how the dogs were raised – e.g. diet or exercise).

It will be several months before we will have analyzed the material on health and behavior and we will not describe the results of the chest white patch which demonstrated what you already know –i.e. that there are large genetic differences in this trait.

We have learned much from the anatomical measurements. We asked you to measure in centimeters because such measurements can be easily compared to other records and because the metric system is a decimal system. Table 1 shows the results for anatomical characters. There is a large range of variation for all traits; and heritabilities (not shown) have told us that there is a genetic basis for this variation in traits 1–4 and 7. Only two characteristics, distance between eyes (5) and muzzle length (6), had low heritability. These apparently were difficult to measure with accuracy given only a measuring tape. The lack of heritability means that we cannot evaluate the "genetic range" of variation, which could be much smaller than the measurements indicate.

For all characters there was a significant difference between males and females. (There was no such difference for white patch, not shown.) We have concluded that there is indeed a great amount of variation within the small gene pool, which is the foundation of the breed.

	1. hock height	2. Elbow height	3. Wither height	4. Tail length	5. Distance Between eyes	6. Muzzle length	7. Ear length
Male range	11cm-21cm	24.5cm-36cm	47.8cm-60.5cm	26cm-44cm	3.5cm-8cm	6.5cm-11cm	4.5cm-15.5cm
Male average	15.7cm	28.7cm	53.3cm	32.8cm	5.5cm	8.7cm	11.3cm
Female range	11cm-18cm	22cm-33cm	43.3cm-58.5cm	24cm-38.5cm	2.8cm-7.5cm	5cm-11.5cm	7cm-15cm
Female average	14.7cm	26.7cm	50cm	30.6cm	5cm	8.2cm	10.7cm
Sex difference	1.04cm	2.03cm	3.34cm	2.19cm	0.5cm	0.48cm	0.65cm

 

All of the measurements were distributed over the population in a normal distribution of the type shown in the figure 1 (below). We have prepared a "Georgie" pedigree(s) of your dog(s), which is included. Dogs in red are ones that have participated in the Georgie Project. Please notify us of any inaccuracy in the pedigree. Your dog's phenotypic measurements are included along with the percentile rank in the distribution of values for his/her sex. We detected possible errors in a few measurements indicated by an *; and these values were excluded from the overall analysis. Table 2 presents the average (mean) values for male and female dogs:

Note that because they are average values they all represent the 50% value in the distribution in figure 1. We have also included a set of values that might occur for a typical dog. Note that some of these values are greater than average, others are lower. We have located these values in the figure.

In addition to the phenotypic measurements and percentile ranking, we have included three other values: COI is the coefficient of inbreeding which represents the % of the genes which were identical in your dog's sire and dam. Tabu represents the percentage of genes in your dog, which are derived from this Algarbiorum ancestor. Taro represents the percentage of genes in your dog derived from this Alvalade ancestor.

Thank you for your cooperation. We look forward to more collaborations in the future.