www.labradorcouncil.com

See above website for more information about silvers, charcoals, and champagnes. The Council for Purebred Labradors is for those of us who love Labs in ALL colors. 

The AKC Stand on Registration of Silver Labradors:

Response of Jack Norton of the AKC on 1/24/00 giving AKC official position on the issue of Silver Labs.

The registry of the American Kennel Club is based on parentage and not the coat color of a member of any breed.

In 1987 the AKC, in corporation with the Labrador Retriever Club of America, conducted an inquiry into the breeding of litters that contained members that were registered as silver. An AKC representative was sent to observe these dogs. The report and color photographs of these dogs were reviewed by AKC staff and representatives of the Labrador Retriever Club of America. Both Parties were satisfied that there was no reason to doubt that the dogs were purebred Labrador Retrievers, however they felt that the dogs were incorrectly registered as silver. Since the breed standard at the time described chocolate as ranging in shade form sedge to chocolate, it was felt that the dogs could more accurately be described as chocolate rather than silver. This remains the current policy of the American Kennel Club.

Jack Norton 
Special Services Dept

Some Falsehoods Regarding the Silver Coloration in Labs

There is no such thing as a silver Labrador.

False. This is really more an argument based on semantics and upon which most conflicts regarding the silver color in the breed are based. Silver Labs do exist here and now, however, history records of the Labrador breed strongly support the conclusion that the silver color was introduced sometime in the mid-history (between the 1940s - 1950s) of the Labrador breed (see below for discussion of origins).   For this and other reasons, the trait is not considered as being representative of the breed.

The silver coat color has been recorded in early writings about Labradors.

False. There is no record of "silver", "gray" or any other color that could be construed as silver between 1878 and 1948 (i.e. the early history of the breed) in the breed stud books. Reportedly, a Norwegian Elkhound cross was performed sometime in the 1940s which coincides with subsequent European reports of "silver" Labradors appearing in some lines during the 1950s and 1960s (see below for more information).

The small gene pool in the early history of the breed had made it difficult to select for silver because "there was never a large enough gene pool of other grays to replicate the color."

False. It is a well established principle of genetics that the smaller the population (i.e. gene pool) the more likely for the offspring of the population to express traits associated with recessive genes.

Genetic analysis has demonstrated that the gene responsible for the silver coloration in Labs is mapped to a site different from the site responsible for the silver color in other breeds of the dog.

False. There is currently no scientific data, either published or preliminary, which has mapped the silver gene locus in Labs. Scientists at the University of California, Berkeley, are, however, currently conducting pedigree analysis on silver bloodlines to determine origins of the silver coloration.

The observation that general breeders of Labradors have not accepted the challenge to disprove the purity of Silver Labradors is confirmation that the bloodlines are pure.

False. General Lab breeders have consulted with geneticists on the feasibility of performing genetic analysis on the silver Labradors.  Current DNA technology is, ironically, too specific a method for assessing genetic relationship between silver Labs and the general Lab population. Although parentage can be determined by DNA testing, there exists too much genetic diversity even between related Labrador bloodlines, which limits the ability of this method to prove or disprove the degree of relatedness between silver Labs and the general Lab population.

Silver puppies born of purebred, AKC registered Labrador parents should be destroyed to preserve the breed.

False. The silver coloration is considered a serious fault in the breed, however, it does not alter the health or disposition of the dog. 

 

Additional information on the silver coloration:

Are there silver Labs?

This is one of the most common questions voiced by the public. Commonly the answer is provided that " the silver Labrador is not a Labrador at all because it is a product of crossbreeding (interbreeding)." This statement may hold both truth and fiction.

Some individuals may argue from the point that up until a few decades ago, the chocolate Lab was, indeed, a minority with many people doubting the genetic integrity of the chocolate coloration. In bench competition, chocolates were often disqualified from competition and even in recent years the chocolates still are considered disparagingly among some Labrador enthusiasts. However, the "liver" (chocolate) coloration appears in most of the retriever breeds and also appears in the original ancestors of the Labrador. In support of this, records of early breedings, around the late 1800s, confirmed that livers were occasionally whelped to black Labs.

Though it is suspected that the yellow coloration was the product of interbreeding, yellow Labradors have been around since the breed was originally accepted as a purebred dog. Therefore, the yellow coloration, though foreign to the original ancestors of the breed (since no yellowed-colored water dogs, only black and occassionally liver-colored, were ever documented as arriving from the breed's country of origin, St. John's) was recognized as an acceptable trait in the Labrador.

So, what of silver? Given the fact that much inbreeding was performed during the early history of the breed because of the small gene-pool, expression of the silver trait would have occurred at least frequently enough for someone to take note of its existence. This was, indeed, the case with the expression of the "black and tan" trait. Early history cites cases of puppies born with tan points (as found in Dobermans, Rottweilers, etc.). This trait was attributed to early interbreeding with Gordon Setters. There is no record, however, of silver Labs or any similar color documented in the stud books spanning the years 1878 to at least 1948 (though other color oddities are documented). This strongly suggests that the silver color is not a color that was present (indigenous) in the early ancestors of the Labrador breed. Therefore, the color must have been introduced sometime after the 1940s. The instances of silver Labs appearing, albeit rarely, in litters from the general population that bear no common ancestors within several or more generations suggests that the gene has been in the population for quite a few decades (This does not necessarily rule out the possibility of more recent interbreeding to purposely achieve or increase frequency of expression of the color). As such, possibilities for the origin of the silver gene may include, but are not limited to the following: 1) a spontaneous gene mutation, or 2) mid-history introduction of the silver gene through interbreeding with a breed carrying the silver gene.

Spontaneous gene mutations occur frequently within any given population. Conceivably a change, most likely in the melanin-stimulating-hormone receptor (Mc1-r) encoded by the Extension Locus (E), could result in a dilution of eumelanin (black/brown pigment) synthesis in Labradors inheriting this mutation and lead to expression of a charcoal or silver coat. Mutations of the Mc1-r have been reported in canines, including the Labrador, and other species (see B/b, E/e, and Beyond: A Detailed Examination of Coat Color Genetics in the Labrador Retriever). Characterization of the Mc1-r in these silver Labs may provide some clues to support or deny spontaneous mutation events as a potential cause for this coloration. To my knowledge, however, no such studies in regard to silver Labs are currently being conducted.

In regard to interbreeding, it is important to understand that purebred development was frequently based upon crossing one breed with a different breed to bring in desirable traits from that different breed. (i.e. introducing "foreign" genes). Through careful breeding programs, early breeders were able to select for the desirable traits of the "foreign" dog while breeding out the other obvious, non-standard traits characteristic to the "foreign" purebred. Mary Roslin-Williams, in her book All About the Labrador, describes a prime example of this, which may have direct implications regarding the origin of the silver Lab phenotype. In her book, she makes reference to a Norwegian Elkhound/ Labrador crossbreeding occurring in the 1940s, as well as to Pointer/Labrador crossbreedings occurring in some field lines. Generally speaking, one may recognize why frequent, widespread crossbreeding, especially in inexperienced hands could cause considerable problems within any breed. However, not all instances of this practice should be viewed negatively. In fact, from a genetic standpoint, there are many positive arguments for the occasional, but controlled use of interbreeding (refer to: Purebred Dog Breeds into the Twenty-First Century -- Achieving Genetic Health for Our Dogs by Dr. J. Jeffrey Bragg). In regard to the Labrador, it is important to recall that at that time, the yellow Labradors were devoid of type, appeared houndy-looking, and had no undercoat to speak of. Crossbreeding of the Norwegian Elkhound offered a quick means of introducing the correct undercoat. Furthermore, the two breeds were similar in terms of structural build. Indeed, this crossbreeding, which may have been one of the key factors leading to improvement of "type" in yellow Labs, may also provide another explanation of how the silver phenotype was introduced into the breed: the silver gene found in the Norwegian Elkhound.

Some purists may be alarmed by this information, however, from a genetic standpoint, the selection and cultivation of "Labrador traits" and the elimination of traits foreign to the breed over subsequent generations has assured the genetic integrity of breed as being "Labrador", even in the presence of such historical crossbreeding. However, one may understand the importance and necessity of a breed standard for ensuring a general consistency among individuals of the breed. As with other traits that are a throw-back to early interbreeding, such as the black-and-tan (attributed to early interbreeding with the Gordon Setter), breeders concerned with maintaining the original attributes of the breed recognize these traits as being associated with the remnants of past crossbreedings.  Additionally, to maintain a general consistency within individuals of the breed,  selection against undesirable traits, whether they are due to "spontaneous mutation" or "foreign" genes introduced by selective crossbreeding, is maintained. It is for this reason that we rarely see evidence of "foreign" genes in the Labs of today.

 

A Matter of Breeding Ethics

Canine genetics is a fascinating area of science. Experimental interbreeding performed by early researchers like Little and Whitney provided many answers to modes of inheritance of many traits including but not limited to coat color. Such purposeful interbreeding was carried out to increase breeders' knowledge and therefore assist them in producing better pure-bred dogs. But what exactly is a better pure-bred dog? Because opinions and tastes may vary widely from one breeder to the next, if left to the individual breeder, one particular breed could become so diversified that members of that breed may look and act nothing like other members. That is why the American Kennel Club, working with the breed parent club, sets the standard for each breed. The standard describes physical and temperament characteristics which are inherently typical of a dog of that particular breed and therefore serves as a guideline for breeders.

As with early breeders who cultivated the traits of the Labrador breed through careful propagation of desirable traits and elimination of faults through selective breeding programs, breeders today follow the breed standard to ensure that the qualities of the breed are preserved. Therefore, at this time, traits such as black and tan, brindling, and silver coloration are considered serious faults of the breed and purposeful selection of these traits for breeding purposes is not recommended. Additionally, the American Kennel Club recognizes only Labradors which are black, chocolate, or yellow. The AKC standard for the Labrador specifically states: "The Labrador retriever coat colors are black, yellow and chocolate. Any other color or combination of colors is a disqualification."

Will silver Labs make good companions?

As with any Lab, temperament and type will depend on the bloodlines of the dog regardless of color. Depending on the breeding lines, a silver Lab can make just as lovely a companion dog as a Lab of any other color. However, as with any Lab that may express an undesirable hereditary trait, a silver lab should be placed in a pet home without registration papers or with Limited Registration to ensure that the fault is not passed to offspring.

A final word on the silver coloration:

The origin of the silver coloration in Labradors remains uncertain at this time. For the AKC to recognize the silver coloration, the parent club would first have to rewrite the standard and vote to accept the silver coloration. For the first of these situations to happen, the silver Lab would have to gain support among a number of its parent club members (as the yellow coloration once had its enthusiastic supporters back in England during the early days of the breed). This scenario is most likely not to happen in the near future. As such, breeders, either established or novice, who may consider breeding for silver will most likely find many doors closed to them in terms of breeding to the best Labrador bloodlines. As such, there are many factors (of which the true origin of the silver is just one) to take into consideration before a breeder or owner should consider the silver colored Lab.


The above information came from http://www.labbies.com/silver.htm

The information below came from http://www.labbies.com/genetics2.htm


The E+ allele: Enter the Potential for a "Gain of Function" Mutation in the "e" allele of the Extension Locus

Color oddities have occurred occasionally throughout the breed history of the Labrador Retriever. Such variations on the typical black, chocolate or yellow coloring have included but are not limited to black-and-tan points, brindling, and silver-casting. It is important to recall that during the early and perhaps mid-history of the breed, interbreeding with other breeds occurred. Crossbreeding to breeds that carry the "wild-type" extension allele (E+) as well as the possibility of a spontaneous mutation resulting in a "gain of function" (also denoted E+) of the Mc1 receptor encoded by the recessive "e" allele, may be possible explanations for color oddities occurring in the breed.

The E+ allele (whether wild-type or "gain of function") would encode a normal, functional Mc1 receptor that would be under greater influence of the alleles at the Agouti (A) locus. Unlike the E allele mutant found normally in Labs that does not require activation by MSH to produce eumelanin (and therefore is always "turned-on" producing black or brown pigment), the Mc1 receptor encoded by E+ would be dependent upon MSH for production of eumelanin (black/brown pigment). Homozygous E+ would allow the effects of the Agouti locus that would otherwise not be seen in a Lab carrying the typical mutant "E" allele (in the typical Lab, effects of Agouti are only seen if the dog is homozygous "e"). Therefore, a Lab that is homozygous E+ would also have to carry As to appear solid black (or chocolate dependent on the allele at the B locus). If a recessive allele (ay, at or as) were the most dominant allele at the A locus, then the Lab would appear either red/yellow (ay or as) or black-and-tan (at). Other Agouti alleles such as aw (which is attributed to producing silver in some breeds) may also be observed.

    Labs having Tan Points

        Early breeding records indicate that a Labrador puppy with tan points on the ears, muzzle, and above the eyes (as found in the Doberman and Rottweiler) would occasionally be whelped to pure-bred Labrador parents. Breeders attributed this to previous interbreeding of Labradors with Gordon Setters during the early history of the breed. Because this trait was considered undesirable as a characteristic of the breed, breeders chose not to breed individuals expressing the trait in hopes of reducing frequency of its expression in future offspring.

        Today, it is recognized that tan points are controlled by the "at" allele of the A locus and that it is recessive to most of the alleles found at the A locus of Labs. Because this allele is recessive, it may be passed through many generations before a breeder is aware that the allele is present. In order for the allele to be expressed, a carrier would have to be bred to another carrier of this same allele and both parents would have to be carriers of the wild-type (E+) Mc1 receptor. This explains the low frequency of expression of this trait in the current Labrador population.

     

    Brindling Effects And "Mosaics" in Labs

        Brindling describes alternating expression of black and red color in the hair throughout the coat. There are several possible causes for this fault that occasionally appears in Labs. One cause may be attributed to the "ebr" allele that controls brindling in many other breeds of dogs. For expression of this trait, both sire and dam would have to carry the mutant "ebr" allele, which is recessive to the "E" allele, but more dominant than the mutant "e" allele for yellow.

        Alternatively, brindling in Labs may be the result of what geneticists call a mosaic. A mosaic indicates differences in the somatic tissue of heterozygotes that come about during mitotic division of somatic cells (recall from above that somatic cells are those that make-up the body). There are two possible ways by which an individual may become a mosaic. The first is called chromosome nondisjunction by which during division into daughter cells, one of the chromosomes fails to separate from its duplicated chromosome. As a result, one daughter cell receives an extra chromosome and the other receives an unpartnered-chromosome.

        The second way that a mosaic may be produced is called chromosome loss by which the chromosome containing the dominant allele gets left behind when the daughter cell's nucleus reconstitutes.

        In either situation described above, the daughter cells of these altered somatic cells will contain the same alterations. As a result, one will observe a mosaic or brindled pattern of normal color mixed with color produced by the altered somatic cells. This condition has been reported in a Lab showing mosaic black and yellow coat color. When this Lab was bred to other Labs of normal coat colors of black, chocolate, or yellow, it was determined that the variation in color was not due to a mutated E locus allele (like the "ebr" allele) because none of the offspring demonstrated this phenotype. Rather, this coat characteristic was attributed to a chromosomal alteration as described above.

        Therefore, the brindling phenotype rarely observed in Labs might be the result of a stable allele mutation (such as the "ebr" allele), or a random somatic chromosome mutation involving the E or B loci.  To view an example of a mosaic occurring as a result of a random somatic chromosome mutation involving the E locus in the Labrador Retriever click here.

     

    Silver Labs

        The silver coat color in Labradors has gained much attention recently and is a very controversial topic (see The Labrador Coat Color Controversy: Do Silver Labs Really Exist?). Reasons for the controversy stem from the lack of information available to trace the origins of this color in the breed as well as the fact that the AKC standard for the Labrador breed does not acknowledge silver as an acceptable color for a Lab. Some breed enthusiasts consider the silver coloration to be a sign of impurity of the bloodline, however, what geneticists have come to understand of recessive alleles is that they may be passed through many generations going undetected, such as the allele for tan points discussed above.

    The range observed in silver coloration suggests that silver occurs through a modifying gene. There have been several possible outcomes observed for the silver Lab:

Black Lab + silver modifier = charcoal gray coat with a "sparkly"-like appearance. Nose: dark gray; Eyes: dark to light gray

Chocolate Lab + silver modifier = "mousy"-brown gray coat. Nose: same as coat; Eyes: yellow to gray-yellow

Yellow Lab + silver modifier = platinum to pale silver (yellow with gray casting). Ears: gray (instead of red-toned); Nose: dark to pale gray; Eyes: dark to pale gray.

        There are several possible explanations for the silver coat color in Labs. The first explanation would attribute this rare color in the breed to the D locus. Recall that the alleles of the D locus modify the color determined by the B locus. Therefore, if a dog is homozygous or heterozygous for black at the B locus, presence of homozygous recessive "d" at the D locus would dilute the black pigment to appear blue. Alternatively, if a dog is homozygous for chocolate at the B locus, presence of homozygous recessive "d" at the D locus would dilute the chocolate pigment to appear silver. The absence of the corresponding "blue" phenotype in the breed, however, would seem to argue against this explanation.

        Another explanation for silver coat color in Labs would attribute this color to the C locus. There is an allele mutant at the C locus that has been determined to cause silver coat color and blue eyes in dogs. The "cb" allele is believed to be a type of albinism. Since alleles at the C locus influence red pigment only, effects of the "cb" allele should only be observed in dogs homozygous "e" at the E locus. Therefore, a silver Lab would not only have to receive the yellow allele from both parents, but also receive the silver allele from both parents (which is recessive to the common "cch" allele).   This allele would explain the silver-toned modification of coat observed in yellow Labs in the presence of the recessive "e" allele, however it would not explain the eumelanin modification in the black or chocolate-based silvers (since the C locus alleles primarily dilute phaeomelanin).

    Likewise, the possibility of a "partial loss of function" mutation that may have occurred in the dominant "E" allele resulting in  muted tones of eumelanin would not explain the modification of phaeomelanin (yellow).

An alternative explanation for explaining the modification of both eumelanin and phaeomelanin again returns to the wild-type/gain-of-function "E+" allele that encodes for a normal functioning Mc1 receptor. If this allele either occurred as a spontaneous mutation or was introduced into the breed through interbreeding, this might explain the modification occurring in all three colors, particularly when one considers the following:

When one traces the pedigrees of some silver Labs, one finds a history of other color oddities occurring in some related bloodlines to the silver Labs. Occurrences of "black-casting" in chocolates, muted chocolate coloration ("card-board box" coloring), as well as the occasional occurrence of black puppies being whelped from two chocolate parents suggests that these "chocolates" were probably not chocolate at all but rather E+ yellows. As such, it is conceivable that the Agouti alleles could produce an intense red pigment resulting in deep red (interpreted as chocolate especially in the absence of "saddling" modifiers) or diluted, muted red (card board box color) due to further modification by the alleles of the C locus). In black Labs, an ayayEE+ geneotype could produce a muted black color (because of the presence of both receptor types) especially if the alleles at the C locus were cch, thus resulting in a deep charcoal, silvery coat appearance. This suggests a possible role of E+ for the silver coloration as well as for a multitude of other coat color variants that occasionally occur in the breed.

     

    White Hair Conditions in Labs

        There are several conditions that can produce white hair in Labs. Some of these conditions are determined by color genes and others may be caused by environmental factors that effect melanin production.

     

    White Spots

        To analyze the reason why some black Labs have only a few, not-easily-seen, white hairs on their chests while others have small white spots, it is best to first picture that all Labs are white--the condition of having no melanin production. The gene loci for color control both the color of the pigment as well as the distribution of melanocytes throughout the body of the Lab. Therefore, in a black Lab, although color is determined by alleles at the B locus, alleles at the A and E loci determine even distribution of the color over the entire surface of the coat. Labs that carry an allele other than "As" at the A locus, have a greater likelihood of expressing more white hairs than those Labs that do carry "As". Therefore, although all Labs should be homozygous for the S allele at the S gene locus, some may still express white hairs on the chest, bottom of the feet and under the arms and groin areas.

     

    Graying In Labs

        Unlike the even distribution of graying observed in some breeds of dogs as they get older, graying in Labs, particularly noticeable in blacks and chocolates, occurs in distinct areas such as the muzzle and paws. Graying may occur as a result of certain genes or as a normal process of aging. In regard to genetic causes, the dominant "G" allele is responsible for causing reduction in the number of melanocytes so that melanin production is diminished. However, in the breeds carrying the graying allele, this effect on pigment is an early event, therefore, it is unlikely that the graying observed in older Labs is a result of this allele.

        Alternatively, the process of aging is also associated with loss of melanocytes and reduction in pigment production. Melanocytes in the muzzle and paws of the Lab may be more prone to the aging process and thus gray more quickly than other locations of the body. Additionally, the reduction in blood circulation and resulting cooling of extremities in geriatric Labs may reduce production of melanin under the control of cold-temperature-unstable tyrosinase enzymes as explained above.

        Other potential causes for white and gray hairs include tissue injury that may destroy the melanocytes in a particular area as well as dietary deficiency of copper, which is required for the production of melanin. The former would appear as localized depigmentation, the latter would appear as an evenly distributed loss of pigmentation.

     

    White "Ring around the Tail" and Two-Toned Appearance in Black Labs

        Helen Warwick reported the former condition occurring in Labs in her book "The Complete Labrador Retriever." In one chapter, she describes that some Labs have white at the base of the hair shafts on the tail. This white color is only discernable when one lifts up the hairs to view the base of the hair shaft closest to the skin. Mrs. Warwick made note that she had observed this condition primarily in Labs of English descent.

        This condition, which is often not limited to the tail, is actually found frequently in both black and chocolate Labs and most likely indicates that the Lab is heterozygous at the Extension (E) Locus ("Ee") and does not carry the As allele at the A Locus.  These Labs also can be observed as having red tones in their coat (occasionally causing a two-toned appearance especially obvious during shedding season). This two-toned appearance is not attributed to chocolate undertones, as some breeders may believe, but is rather due to the production of the red pigment, phaeomelanin, in "e" (yellow) carriers that also carry "ay-" or "as-".As a result, both Mc1 receptor types will be present in the dog: one that continuously makes eumelanin and one that will only make phaeomelanin.   The receptor encoded by "E" for production of black/brown is more efficient, however, some phaeomelanin will be produced by the "e" receptor.   Since the recessive Agouti alleles control synthesis of phaeomelanin during particular times of hair growth (with phaeomelanin produced only during the mid to late portion of hair growth), phaeomelanin will only be seen at the base of the hair shaft. If these Labs also carry the "cch" as the dominant allele at the C locus, the red pigment will be diluted allowing the base of the hair shaft to appear very light in color. 

     

    Final Remarks

    To date, much of the knowledge pertaining to coat color inheritance is very limited with most of the information either derived from small canine studies or extrapolated from other species. Therefore, much of the data is still open to interpretation. In light of this, this article presents the current consensus for inheritance of coat color based on the previous findings and interpretations of those in the field. It is anticipated, however, that our knowledge of canine genetics will be greatly enhanced over the next several years as scientists conducting the "Canine Genome Project" draw closer to their goal of mapping specific genes to their chromosomes in the canine.

    genome.gif (4280 bytes)

    References

    Newton JM, et al. Melanocortin 1 receptor variation in the domestic dog. Mammalian Genome, 11:24-30, 2000.

    Everts RE, Rothuizen J, van Oost BA. Identification of a premature stop codon in the melanocyte-stimulating hormone receptor gene (MC1R) in Labrador and Golden retrievers with yellow coat colour. Anim Genet, 31:194-9, 2000.

    Lu D, et al. Agouti protein is an antagonist of the melanocyte-stimulating-hormone receptor. Nature, 371:799-802, 1994.

    Vage DI, et al. Molecular and pharmacological characterization of dominant black coat color in sheep.  Mammalian Genome, 10:39-43, 1999.

    Vrieling H, et al. Differences in dorsal and ventral pigmentation result from regional expression of the mouse agouti gene. Proc.Natl.Acad.Sci. USA, 91:5667-5671, 1994.

    Cornell University Animal Health Newsletter, "Coat Color Inheritance," Volume 14, Number 11, January 1997.

    Sponenberg DP and Bigelow BJ.  An extension locus mosaic Labrador retriever dog. Hered.,  78:406, 1986.

    Suzuki, D.T., Griffiths, A.J.F, and Lewontin, R.C. (eds.), An Introduction to Genetic Analysis,San Francisco, W.H. Freeman and Co., 1981.

    Fitzpatrick, T.B., Szabo, G., Wick, M.M., Biochemistry and Physiology of Melanin Pigmentation, 1981.

    Helen Warwick, The New Complete Labrador Retriever, New York, Howell Book House, Inc., 1986.

    Helen Warwick, The Complete Labrador Retriever, New York, Howell Book House, Inc.,1965.

    Leon F. Whitney, DVM, How To Breed Dogs, New York, Howell Book House, Inc., 1972.

    Little, C.C., The Inheritance of Coat Color in Dogs.  Comstock Publishing Associates, Ithaca, 1957.

    Willis, M.B. Genetics of the Dog. MacMillan Publishing Co., New York, 1989.