Created by Karl Pearson
1913
Description:

‘In the year 1906 the late Edward Nettleship and I began a lengthy inquiry into albinism in man. Before long we took into partnership C.E. Usher, whose large contributions to our series of pedigrees and observational work on all forms of leucosis more than entitled him to full responsibility for our joint publication. Nettleship’s splendid power of following up details of family history, his relative leisure, his patient enthusiasm, and above all his wide ophthalmological knowledge provided what I personally largely lacked, and his recent death has been a grave loss to our joint work. Luckily in my remaining colleague, Mr Usher, most of these features are still present, combined with a youth and vigour, which is of endless profit to our joint work.

Early in the course of our researches, we found that the difficulties arising in dealing with man for the several problems of albinism were very great owing to the extreme rarity of material for microscopic examination and we were forced for many of our problems to turn to other types of life. Albinism is relatively rare in the [1-2] dog, but fortunately for our work Nettleship in 1907 was able to procure three albino dogs Tong I, Jack and Jill, and later these were added to by the purchase of two others Beenie and Spook, while certain number of other albinos have been offered to me of recent years. I have had no necessity for them because we have bred something like 50 or 60 albinos from our foundation stock. All these dogs are pure bred Pekingese, but as they differ in other features, but pigment from the normal dog, we term the whole breed of pure-bred Pekingese albino dogs Dondos. Every Dondo that has been brought to me or that I have received any report of – and Dondos occur sporadically among the Pekingese – is ultimately descended from a certain pair Ah Cum and Goodwood Meh. Five Pekingese – the first to come to England – arrived after the looting of the Imperial Palace in 1860 – one ‘Sooty’ went to Queen Victoria [note: ‘the Queen, I mean’], was kept in the Royal Kennels & never bred from. Schlorff and Hytien went to Lord John Haye, and formed the basis of his extensive kennel. He has bred once a white [2-3] dog with dark eyes & black nose, but never had an albino. The other pair of the 1860 batch Guh & Meh were rich chestnut brown with black masks & points. They were kept going by direct descent at Goodwood till the early nineties without record of albinism. Then 1896, Ah Cum & Mimosa were brought from the Imperial Palace at Pekin by Mrs Douglas Murray, and from the mating of Ah Cum with Goodwood Meh – a descendent of the 1860 Meh – or from his supposed sister Mimosa – all the albinos that I have been able to trace descended. The only rule to avoid an albino Pekingese appearing sporadically in your kennel is to avoid the Ah Cum & Meh stock, or imported Pekingese, who are likely to come from their ancestry. I have not been able to convince myself that true albinos have appeared in China. They are cream white, or yellow dogs with black or liver colour noses, but if absolute albinos have appeared, they are probably killed, much as human albinos in China till quite recently appear to have been disposed of or outlawed. One of the Chinese names for albinos is Yang ren – ‘foreign devil’ – a name, I promptly gave to my early & most perfect Dondos.

Now albino has been defined as “an organism in which there is complete absence of pigmentation, other than the blood and bile pigments.”

It is one of the evils of science that definitions are usually made by text-book writing or those who have not very fully examined the material bearing on the subject. Albinism may be defined as a lack of full normal pigmentation, but an absolute lack of melanin pigmentation would exclude something like 95% of albino birds, nearly all human albinos, and a very great number of albino mammals. How many it is impossible to say, because but little adequate microscopic examination has been made. Even today but little is known about the pigmentation of the internal organs of albinos. Our albino dogs are free of all internal melanin pigment, a deaf white cat showed absence of internal pigment granules in all the internal organs where pigment is usually found. On the other hand an albino rabbit and a rat were reported to have internal pigment, while a second albino rat had none. As a result the pure bred Dondo has no pigment granules in the hair of the coat, and this statement is consistent with quite a considerable macroscopic colour towards cream or yellow. Usually the eye is largely devoid of pigment, but there is always some of the epiblastic layers of iris and ciliary body. Further, in the coats of some Dondos as puppies a few pigment granules will be found, which sometimes dis-appear with the first or second shedding of the coat, but in some Dondos there is a permanent slight pigmentation of the hair. Now all the points can be paralleled from human albinism. Of course very few sections of the human albinotic eye have been microscopically examined, but in such that have it is the rarest exception not to find some granular pigment; it occurs almost invariably in the eyes of adult human albinos. One of the only cases, I know, of no granules being discovered, occurred in the case of a very noteworthy albino, but there we could only examine a small portion of the iris removed in an operation for cataract, and it is impossible to argue from an absence here to an absence everywhere. Even in the most complete human albinos we have been sometimes able to extract single hairs from the scalp well pigmented, and one perfect ‘clinical’ albino of our record has a jet black lock of hair at the back of his head. In every feature from approximately complete albinism to very imperfect albinism and even to piebaldism, the parallelism between albinotic man & albinotic dog can be followed up. Perhaps the general “albinotic facies” is the most striking feature of resemblance in man & dog.

I would first familiarise you with the appearance of the Dondo. Here are photographs. You will see a living specimen outside. The first question is, does the Dondo breed true? The real answer to that question involves a treatise. In a certain sense it does breed true. That is to say there is something albinistic – some degree of albinism about the offspring, but there can be very considerable range of variation in the albinism, and this range not only comes from different albinos, but from breeding with the same pair. A recent review in ‘Nature’ of our monograph on albinism was very contemptuous of our work because we had not divided up our albinos into Mendelian classes, and suggested that the chief value of our work would be the provision it made for somebody else to carry out such analysis. I am fully aware that Mendelism having started with the broad categories: albino and not albino, now tells us that thirteen types of albino rats exist, and that with rabbits and mice there must be a still larger number. Perhaps in men and dogs they will assert 20 or 50 types of albinism. I merely outbid them and assert that if they come to the hair, the eyes & the skin in men & in dogs and examine them microscopically they will find an infinity of types, every individual is practically a type, and rigid discussions between grades of pigmentation do not exist. We have bred over 50 pure-bred Dondos. They do breed true to albinism in general, but exceptions arise. To begin with the eyes after death examined microscopically do show considerable range of pigmentation, precisely as occurs in the eyes of the albinos of dark races & probably in a less marked way in the eyes of European albinos – certainly in those of Armenia and Syria. You cannot divide eyes up as some Mendelians have done into true blue lacking in anterior pigment and not true blue eyes possessing anterior pigment. Eyes passed - far more carefully than is obviously done in the published Mendelian researches – as true blue macroscopically, show usually some colour pigment on microscopic examination, while the stroma lying between the anterior and posterior surfaces is dotted in greater or letter extent with pigment granules. There is every grade of pigmentation from a markedly loaded anterior surface to one passed as without anterior pigment in which a few granules will probably be discernable with the microscope. Just as the Mendelian category – absence of pigment – fails with eye, so it fails with coat. Marked aberrations are rare, but albinism does not breed true, if by that is meant its absolute like. Jack & Jill have given us very white coated dogs like Patty, but they have also given us cream coloured albinos, down to brown – look at the coats exhibited – and in one rare instance, they gave us Fe a piebald with imperfectly albinotic eyes.

In the above I am speaking of pure-bred albinos, but with extracted albinos I get wider deviations, and have almost established a chocolate albino. In looking through [8-9] the late Professor Weldon’s mice data, I found a record, without comment, of two extracted albino mice producing offspring which could not be called albinotic. I have not ventured to mention that result till tonight, because we have so frequently been told that albino & albino will only give albino, & in the absence of comment I was not sure how far Weldon had himself realised the gravity of the result. But when I have learnt what can happen with extracted albinos in dogs and seen what happens in man, I have little doubt of the authenticity of Weldon’s record. All albinism as a rule is imperfect, especially in the extracted form, and imperfect albinos can give rise to albinos so imperfect, that you are doubtful whether they are to be called albino or not.

Having proved that these pure-bred Dondos formed a breed which would go on with rare exceptions perpetuating itself, we next looked for a type of dog of somewhat the same size, which was in marked contrast for purposes of hybridisation. We chose the pure bred black Pomeranian. It is fairly high from the ground, has a markedly different [9-10] leg shape, a wholly different muzzle, ear, and bark; its psychical characters are wholly different too. Black Poms with the largest amount of black were selected. Olga’s parents were black, her four grandparents, her eight great grandparents, her 16 great-great grandparents and 28 out of her 32 great-great-great grandparents; [30] of the 32 were known and 2 unknown. I would draw your attention to the difference between the muzzle of Olga & the normal Pekingese. The result of crossing the black Pom & the albino Pek is the second type of dog I mention to you tonight, which we term the Pompek. The Pompek is a black dog generally with a white shirtfront. In a few cases the white shirtfront dwindles to a few white hairs. It is rare that such white hairs are not findable. We have had 25 such Pompeks. The muzzle of the Pompek we at first thought Pekingese, but when we compared it with the Pekingese we thought it Pomeranian. When we actually measured the heads of our dogs we found it a blend. There was no dominance of head form in any of its characters. That [10-11] is a strong argument for measurement, rather than mere impression in classifying into Mendelian categories! – But is there dominance of coat-colour? The Pomeranian has no white shirt front, and at least 19 of our 25 dogs have marked white shirt fronts. There is dominance only in the sense that there is more black than white. But now comes a remarkable fact, two out of the 25 Pompeks are not black at all, but they have a white shirt front, but the coat is uniform chocolate. Further, the skin of the paws is not black but liver colour, the noses and lips are liver colour. The eyes of one examined microscopically show faint brown pigmentation only on the fundus and far less iris pigmentation than is the case of the usual Pompek. The surviving chocolate dog examined with the ophthalmoscope showed a fundus obviously paler than that of the normal Pompek. There was red reflex[? – i.e. reflection?] from the eyes of both these Pompeks. The eye was imperfectly albinotic although unequal to that of the Dondo in extent of albinism. In short these chocolate dogs in coat & skin & eyes show a blend of albino & normal black Pomeranian. Two other dogs in this litter are rather dark red, than black, & only one is a normal black & white Pompek. [11-12] Prince crossed with another albino ‘Spook’ gave two normal black & white Pompeks, one very dark red & two Anna & Douglas very hard to describe as red or black described as black when puppies, but red brown now. There is no universal law of dominance at all and we must either assert blending or segregation to occur in the first generation. Unfortunately so far only one male of this type has survived and we cannot yet say what happens, when the chocolate or red Pompek is crossed with the same class.

We are now in a position to cross the Pompek with the Pompek or with the Dondo. I take the latter first. If albinism were a simple character, we should anticipate nothing but Pompeks & Dondos to result. Pompeks do result, though often of a rather rusty black, Dondos result, but often with very imperfect albinism. Dondo & Dondo pups, at birth are either cream white or café au lait, the distinction is quite marked but it very largely passes away in the first year. But in the case of Pompek crossed with Dondo, this café au lait type, or scraped chocolate type can remain permanently, and give a very remarkable new type with albinotic eyes, pink nose and this coloured coat. I have now bred such [illeg.] dogs. Dams from first generation Pompek & pure Dondo, [12-13] are of a litter of which the others were two pure albinos, and there were no Pompeks; and the other three from crossing a pure Dondo Wang, with a Pompek extracted from Pompek & Dondo. I hope these café au lait albinos may perpetuate themselves, but crosses are not possible as yet, as we have only yearling dogs. In the hope that we may establish a permanent breed of these semi-Dondos, I have given them an independent name to the memory of the first great student of albinism, the Belgian [sic – Eduardo Cornaz appears to have been Swiss] ophthalmologist, and call them Cornaz spaniels.

But the 2nd generation Pompek, the 2nd generation Dondo & Cornaz spaniel are not the only types which appear when Dondo is crossed with Pompek. Brindle red dogs have appeared in this generation. These brindle red dogs when crossed inter-se give almost entirely brindle reds, the exception being about one albino in seven or eight. I have bred about 17 of these red brindles now & hope to get rid of the occasional albinism. They are a most delightful type of dog and I find them in great demand. I have named them the Galton spaniel. Thus the [13-14] Galton spaniel arises either from crossing Pompeck with Dondo, or from crossing Galton spaniels inter se. I have also got it by crossing Galton Spaniel with Dondo in which case it seems to me a more beautiful dog than when obtained in the other two ways, as it tends to be more pure red & less brindle. The Galton spaniel is a most charming and, I think, if it be possible, deserves to be established as a separate breed. At present the head wants steadily[?] selecting as it looks, what it is, a blend. As in the case of Pomeranian crossed by Dondo, there have been exceptions to the rule that Dondo crossed by Pompek, gives Pompek, Dondo, Cornaz spaniel or Galton spaniel. My colleague Mr Usher has obtained one very remarkable litter consisting of two Pompeks, two black & white dogs, genuine piebalds with dark eyes, and a café au lait or lilac skewbald with albinotic eyes. All died as puppies except one Pompek & it has not yet been found possible to repeat the experience. The relation between piebaldism & albinism is far closer than is usually admitted by Mendelians. Our crosses of Pompek with Pompek have so far not been very successful. I have made [14-15] myself four such crosses, two were infertile the other two produced six puppies, which consisted of two Pompeks, two albinos (extracted), and two red dogs or Galton spaniels. Mr Usher has made four such matings leading to two extracted albinos, one fairly complete the other imperfect, to one [illeg.] dog, which we may claim as a Galton spaniel, to six Pompeks, and to one ‘chocolate’ dog with semi-albinotic eyes, which I imagine though I have not yet seen the coat corresponds to a Cornaz spaniel.

At present there has been no crossing of a Pompek with a black Pomeranian. I have crossed albino from Pompek & Dondo with a pure bred Dondo. I obtained five albinos, all with somewhat impure coats streaked with dorsal[?] red. They can be distinguished at once from pure bred Dondos by their ugly heads & very often their clumsy size. I have had recently a litter from two albinos of coloured ancestry. The mother was from a Pompek and Dondo, and the father from a Galton spaniel and a Dondo. The offspring are all albinos but so far as can be judged at present - they are only a few days old – of a very imperfect kind. In fact they are all possibly Cornaz spaniels. It must not for a moment be supposed that the types I have discussed are definitely & rigidly defined classes. They certainly are not. The Dondo gives crossed by itself a considerable variety of albino offspring, the true Dondo, the lilac albino, and the lilac skewbald. The extracted albinos range from Dondos white, through almost all phases of streaky red or lilac up to the ‘scraped’ chocolate of the Cornaz spaniel. The Galton spaniel, ranges from the darkest brindle to almost a golden colour; there are dogs we hardly know whether to class as black, rusty black or darkest red. There is a practical continuity of coat colour with lumps at particular grades, those I have indicated with the names I have introduced to you tonight. What happens in coat-colour, happens in eye pigmentation also. The microscopic examination of sections of the eye shows an enormous range not only of variety but quantity of pigmentation. As for head shape the chief characters blend and neither dominate, nor segregate in our experience. Thus far we have not deduced any black [15-16] dogs or red dogs which exhibit pure dominance, there is always an occasional albino turning up. But the albinism tends to be less & less perfect, and, I believe, it will be possible ultimately to obtain almost any grade of albinism. I have at present a bitch with one normal and one albinotic eye, and a brown skewbald coat, Choo Ko, or ‘Wan eye’. Mated with her own father, however, she has not yet perpetuated her hetero-chromia, but given me three skewbald puppies, one of whom Ming Wang, I show you tonight. He has imperfectly pigmented mucous membrane – the liver colour. Choo Ko is like our Dondos a piebald Pekingese directly descended from Ah Cum & Goodwood Meh on one side.

My firm conviction is that if you can obtain in any species with a normal black or dull colouring, an absolute or imperfect albino, then you can secure every variety of colour by judicious mating. The mating of an albino and a normally pigmented individual leads to a general pigmentation break-up and almost every variety of pigmentation will occur. In almost every case of dark race piebald [16-17] that I have been able fully to investigate, you ultimately reach negro and white, negro, and Indian, Chinese, and Philippine crosses. The piebald appears not in the first cross but in later generations. There is a very close parallelism between men & dogs; in human albino families we find a far larger percentage of red hair than in the general population, often of a very brilliant type; the remarkable golden blonds which occur in the piebald in negro albinotic stocks, - an experience far beyond the range of random distribution of such rare anomalies – all emphasise the relation of albinism in man to albinism in dogs. I am much inclined to think that the white man as well as the copper-coloured races, have been deduced by relation of imperfect varieties of albinism from a primitive dark-skinned race.

It has been said by a distinguished Mendelian that the greyhound is an “incorrigible hybrid”, and cannot be used for demonstrating [17-18] hereditary properties. I have no doubt that I shall be told that our Pekingese & our Pomeranians are also incorrigible hybrids. If so, I think we must also include man as an incorrigible hybrid. If this be so, then I reply that I am personally alone interested in the study of heredity in incorrigible hybrids.

The moral of [the] lecture tonight to[sic] this: we started with two types of dogs which bred relatively true to head-shape and coat-colour. We have reached head shapes of almost every variety, and wide ranges of coat-colour; some of these appear to fall into classes, which again breed, not absolutely, but relatively true. As far as I have experience many of the characters thus reached are not characteristic of either original breed. The general disturbance due to the cross has produced fundamental changes in the germ plasm, which, perhaps, some day the cytologist will explain, as he is starting to do with plant-hybrids. I find nothing to confirm the view that hybridisation is a mere shuffling of definite unit characters. Yet that Mendelian view has been preached by anthropologists, and is being now taught by sociologists wholly regardless of the slender evidence on which it is based. What happens asks Dr Fischer of Freiburg after race crossings?

“Does a new race arise? Or a mixed race with blended or new characters? Does one race more or less dominate the other?”

Dr Fischer asserts that the last question has been answered by Bean, Salaman & Davenport for the dominance in hair, eye & facial characters in man.

“Again,” he writes “the answer to the first question has been given by the thousands of hybridisation experiments of botanists & zoologists – no new race is to be expected. Characters always separate out again according to the Mendelian rules, the unit characters are always found in thousandfold combinations, alongside each other. - [..]. characters are persistent, an immense number of combinations arise, but no new race... Von Luschau was the first to emphasise the process, he pointed out how the old types always reappear, types which existed in a country thousands of years ago. Race mixture [19-20] leads, he said, to ‘sorting out’ – (“Rassenmischung zu einer Entwischung”). Today, when we know Mendelian segregation this is intelligible without any further explanation.”

So far Dr Fischer.

According to this view man for thousands of years has consisted of the same component unit characters, endlessly shuffled kaleidoscope fashion. Every character in man as we know him now either existed in Palaeanthropus [sic], or in one of his contemporaries.

To take this view is to destroy the whole philosophical basis of Darwinism. The gradual evolution of man by the selection of small variations suited to his changing environment. Neither theologian nor metaphysician have shaken the foundations of Darwinism in the past to anything like the same extent, as those who assert the absolute true [sic – i.e. truth] of the doctrine of unit characters. The day is not too far distant when we must reject the doctrine of unchangeable unit characters or give up any consistent form of Darwinian evolution. The Biometric Laboratory does not criticise Mendelism because it is supported by those, who have not realised the nature of Galtonian methods, [and wish] to destroy their application to heredity. Our criticism arises from deeper sources. Where we have been able to investigate ourselves, as I have shown you in the case of dogs tonight, there are no such rigid classes as those demanded by the idea of a unit-character, the ‘true blue eye’, and the albinotic eye have really every grade of pigmentation, the piebald has every variation in extent of piebaldism, and these various grades are hereditary. Dominance is not a universal rule of inheritance, there is segregation in the first generation as in the second. There is nothing to lead us to suppose that all changes are due to a shuffling of persistent ‘unit’ characters, and that if new characters have appeared in the course of evolution they are due to an inexplicable ‘mutation.’ The moral of my lecture tonight is twofold – there is still some hope for a philosophical theory of Darwinian evolution; and I urge that those who are rushing forward and describing epilepsy, [21-22] mental defect, insanity, to say nothing of shyness, and indolence, as ’unit characters,’ are going far beyond what the data permit: and their teaching may involve grave social consequences, when we apply it to legislation and advise on the basis of it marriage or restraint from marriage. Albinism was a ‘unit character’ not many years ago, now we are told there are 15 to 20 types of it & perhaps more. ‘Mental defect’ is a ‘unit character’ just now, there will be 15 to 20 types of it, no doubt, in a few years. I appeal for delay in judgement, and [that] the public should not accept without being cautioned, a theory of heredity, which not only reverses all philosophical Darwinism, but which does not stand the analysis of measuring rod and microscope when they are applied to it.’