skip to content

Darwin Correspondence Project

From Alpheus Hyatt   [late] November 18721

Cannstatt bei Stuttgart

Nov 1872

Charles Darwin

Dear Sir

The letter with which you have honoured me bearing the date of Oct 10 has just reached here after a voyage to America and back.2

I have long had it in mind to write you upon the subject of which you speak, but have been prevented by a very natural feeling of distrust in the worthiness and truth of the views which I had to present

There is certainly no occasion to apologize for not having quoted my paper. The law of acceleration and retardation of development was therein used to explain the appearance of other phenomena and might, as it did in nearly all cases, easily escape notice.3

My relations with Prof. Cope are of the most friendly character; and although fortunate in publishing a few months ahead, I consider, that this gives me no right to claim anything beyond such an amount of participation in the discovery, if it may be so called as the thoroughness and worth of my work entitles me to.4

Prof Agassiz5 some thirteen years back introduced me to the study of the Ammonites and entrusted to my care his truly immense collection of these fossils the value of which I appreciate more highly now that I have seen those of Wurtenburg.6 Your time is doubtless very valuable and therefore I have taken the liberty of reducing four pages of irrelevant matter to this one half sheet, for which I beg to apologize.7

My letter will doubtless prove long and tedious, nor am I at all satisfied with the way in which the explanations are given but it is the best I can give at present without delaying my reply until the next week.

The collections which I have studied, it will be remembered are fossils collected without special reference to the very minute subdivisions such as the subdivisions of the Lower or Middle Lias as made by the German authors especially Quenstedt and Oppel,8 but pretty well defined for the larger divisions in which the species are also well defined. The condition of the collections as regards names etc was chaotic, localities alone, with some few exceptions, accurate. To put this in order they were first arranged according to their adult characteristics. This proving unsatisfactory I determined to test thoroughly the theory of evolution by following out the developmental history of each species and placing, them within their formations, Middle or Upper lias Oolite9 or so, according to the extent to which they represented each others characteristics. Thus an adult of simple structure being taken as the starting point which we will call -a- another species which was -a- in its young stage and became -b- in the adult was placed above it in the zoological series. By this process I presently found that -a- then -ab- and -abc-, c representing the adult stage, were very often found; but that practically after passing these two or three stages it did not often happen that species was found which was abc in the young and then became d in the adult.

But on the other hand I very frequently found one in which, while it was -a- in the young, skipped the stages -b- or -c- and became d while still quite young. Then sometimes, though more rarely, a species would be found belonging to the same series which would be -a- in the young and with a very faint and fleeting resemblance to -d- at a later stage pass immediately while still quite young to the more advanced characteristics represented by -e-, and hold these as its specific characteristics until old age destroyed them. This skipping is the highest exemplification or rather manifestation of acceleration in development.10 In alluding to the history of diseases and inheritance of characteristics, you in your Origin of Species allude to the ordinary manifestation of acceleration, when you speak of the tendency of diseases or characteristics to appear at younger periods in the life of the child than of it’s parents.11 This according to my observations is a law or rather mode of development which is applicable to all characteristics, and in this way it is possible to explain why the young of later occurring animals are like the adult stages of those which preceded them in time. If I am not mistaken you have intimated something of this sort also in your first edition, but I have not been able to find it lately.12 Of course this is a very normal condition of affairs when a series can be followed in this way beginning with species a, then going through species ab to abc, then abd or acd and then ade or simply ae as it sometimes comes. Very often the acceleration takes place in two closely connected series thus.


in which one series goes on very regularly while another lateral offshoot of a- becomes d in the adult. This is an actual case which can be plainly shown with the specimens in hand and has been verified in the collections here. Retardation is entirely Prof Cope’s idea but I think also easily traceable.13 It is the opponent of acceleration so to speak on the opposite or negative of that mode of development. Thus series may occur in which either in size or characteristics they return to former characteristics, but a better discussion of this point you will find in the little treatise which I send by the same mail as this letter, “On Reversions among the Ammonites”14   Since coming here I have been endeavouring to verify the series which I have made and have found that, though mistakes are many and often very serious, still in the main the parallelism of a series made up according to the development of the species alone and of series made up according to development and chronological sequence is exact whenever the material is sufficient. You will be doubtless pleased to learn also that the deficiencies in the Geological Record are as apparent between the smaller beds such as the Tuberculatusbett, Bucklandibett, et al of Oppel as between the larger formations.15 Neverthless this record is much more complete and reliable than there was any reason to expect theoretically and often with the aid of development we can bridge over the gaps and make complete series. In my paper in the Memoirs Bot. Soc. Nat. Hist I used the accelerated inheritance of characteristics to explain the degradation of series or abnormal forms or degraded forms rather which often cap series which pass through several formations.16 These forms are often like those with which a given series begins thus the Baculite is like the Orthoceratite or the species Collenotic or pseudostellare (a new species) in many of it’s characteristics resemble the species psilonotus, the former ending the series of Arietidæ and the latter beginning the same, the former in the Obtususbett and the latter in the Psilonotubett of the Lower Lias.17 Everywhere in complete series which develop according the accelerated mode this is found and also in others.

This resemblance is a double one. These terminal forms are at the same time the highest of their series in their organization and development and yet like the most immature in many characteristics. Again these terminal forms have not only these resemblances but they also resemble the old age of earlier species of their own series.


Here for example we have in the first rank three forms adult, second young of the same and in the third the extreme old age of each, all in the Arietidæ and though diagrams in the main strictly correct sections tranversely through the whorls. The first thing to be observed is that the first is psilonotus occurring lowest the next is the Bucklandi (of Germans not English) or Brookii of Germans and the third is either the new species pseudostellare or the older one Collenotii of D’Orbigny.18 The first is lowest the next about the middle and the third higher up chronologically in the Lower Lias. Oppen and Fraas19 and Quenstedt all agree upon this point and the collections here show that they are right. The development of the form the form of the adult, and the smoothness of the shell in Nos 1 & 3 and the great difference shown by the middle form no 2 which is keeled on the abdomen with deep channels and has tuberculated ribs on the sides. Compare this with the development of No 2 and we find that the adult of No 3 is like the old of No 2. Thus not only is adult No 3 like adult of no 1 but it is also like the old of no 2. I have found, or thought I have, that this resemblance was due to the direct inheritance of degradational characteristics, not as an unusual but as a usual mode of development, among the higher or rather later occurring species of different series. Thus between 2 and three, there are a number of species and in these we find that if we represent the characteristics of No 2 by letters thus a=young d=adult and x=old age characteristics of shell that the series may be thus given in a rough way.

adx=No 2. adex–aex–ax.

Thus the series as a whole would thus be given.


From a to ad we have a series steadily increasing or adding in each adult something new to the series and then we have at the same time another series arising later which begins a step in advance of where the first ended but in which the acceleration of the development of the characteristics take place differently. Thus it is first regular repeating the adult, a-d-characteristics of adx in the young of adex and adding e for the adult stage, then the characteristics are still farther accelerated and e appears in the young of aex, but x here no longer represents the old age as formerly. It represents the adult and finally we have a species which is only a x. One in which even e is eliminated or skipped in the development. There are many serious questions which ought to be answered here but they would complicate my statements too much. One however is too important to pass by. Of course I do not mean to claim that old age characteristics are inherited but merely that there is an exact parallelism between degradational characteristics of all kinds and that the degradation which occurs in an old individual taken from the middle of a series may be compared precisely with the degradational characteristics which usually characterize the terminal as later occurring members of complete series. To carry this a little farther. The Orthoceratite in the adult differs but little from it’s own young and however large it may grow has no observable old or degradational characteristics. The ornamented Ammonites and Nautili alone have an adult which is very distinct from the young and they alone, also, have observable degradational senile metamorphoses. Again the Baculite is like the Orthoceratite but little changed by growth either in form external ornaments, or septa, the latter having only six lobes as in the young. Compare this with 1, 2, 3, here we have the same, homogeneity in the beginning, heterogeneity in the middle, and homogeneity again at the end. Does not this compare in the most remarkable manner with the development of any one individual taken from the middle of the series such as No 2. Homogeneous or simple in the young, heterogeneous with ornaments etc in the adult and again homogeneous or devoid of ornaments and smooth again in the old. D Orbigny has noticed this resemblance of the old age of the ammonite shell to it’s own young and in some of them it is most remarkably close. Therefore I did not hesitate to draw a parallel between the life of an individual and the life of the group to which it belongs. The whole series of Ammonoid forms beginning with their Orthoceratitic ancestors mount to the zenith of their development as the Jura and die out in the Cretaceous jus as an individual from the ornamented Jura forms grows to the adult condition and then dies by a series of retrograde, degradational metamorphoses in old age. This exceeding rough statement needs hundreds of qualifications, since I have here expressed myself as if the Cephalopods were one series whereas they are divisible into a great many and these again into others. Tracing out these is a most difficult task. One illustration I believe will carry my meaning better than any other that of a large wave covered with an inexplicable entanglment of smaller waves and yet all setting one way with the wind mounting up on one side of the great wave only to fall again on the other. And now I come to what is most interesting to me. How does all this agree with the Theory of Natural Selection? For years past I have been of the opinion that the regularity with which these series followed out a given line of characteristics in their progress was irreconcilable with your law and, though much more doubtful now, cannot get a clear idea of how Natural Selection can bring about such a series of phenomena or on the other hand how these, as I formerly thought, show some higher and more comprehensive law. They seem to me as you intimate in your letter to be not very fitly expressed by the word “law”20

I think we ought to use the word mode and call it the mode of development through the acceleration and retardation of development. It appears to me that I have made a slight step forward in my paper on Reversions but you are the only judge of this. How can Natural Selection account for the fact, for fact it is according to observations a hundred times repeated upon the Arietidæ and other Amm. that the last of the series have these peculiar degradational characteristics and yet are in other respects the highest of their series? How can it also account for the limitation of the forms which seems to be implied by these observations. Thus we can say with great assurance what will be the coming forms of a series from the first two or three, especially if the old age is shown in one of them? Twice I have been able to verify this. Once in a series


when none other followed I wrote in my manuscript that the next step must be


and a short time after found it in a specimen, which I had always supposed must belong to another series but upon changing thoroughly and examining young and septa traced it to this one.

Another was in the case of the German Bucklandi whose old age I had never seen but predicted what this must be from the adult characteristics of pseudostellare and Collenotii. It does not follow that a series is regular, it may be reversionary or apparently stationary but the result of the whole is normal and symmetrical. Thus among the Arietidæ


they must succeed each other so as long as the series develops normally and then


so if they continue to live longer. Nothing else seems possible. Follow up from the Arietidæ through the Falciferi and Clyperform and others, it is the same.21 Take other large series and the same symmetry prevails with different forms. Take an individual out of the middle of the series and again in a complete life we have


the same formula when in the Arietidæ etc. Can Natural Selection explain this?. This question is perpetually arising in my mind and sometimes it seems as if the solution was attainable. Thus; lately I have visited the famous Steinheim locality which Hilgendorf has so thoroughly worked up. Doubtless you have remarked this paper since it has such an important bearing upon the theory of evolution.22 The formations in the sand pits containing Paludina multiformis23 are merely the successive layers formed at the bottom of a freshwater lake, evidently continuous or nearly continuous deposits, with only such interruptions in the zoological record as might have been occasioned by the migrations of the shells from one point to another according as the deposits were favorable or unfavorable for their life. These interruptions are mostly found where clay has been heavily cast down from the water above but even in these shells are occasionally present. Thus the record is as complete probably as we shall ever find it and the different varieties are nearly all traceable from one to another without break. If you consult Hildendorfs plates you will remark here the same law which characterizes the Ammonites.24 Namely the resemblances which exist between the first and last forms of the large central series and the peculiar degradational characteristics also of the last or upper members of this series, or as revertions especially. Variety Trochiformis has an old age, that much I have seen already but have not yet studied the relations of this stage to the reverters and supremus varieties.25 I think however that in this locality it is possible to find something definite something which may possibly lead to the reconciliation of the facts, which I present above, to the law of Natural Selection. Thus it may be easily seen in the pits that the surrounding circumstances were continually growing more and more favorable up to the Trochiformis bed and then that they became less and less favorable subsequently. The prevalence of clay underneath and above the Trochiformis appears to indicate this and to account for the gradual increase and diminution in the size of the varieties. The inheritance of degradational characteristics on the part of reverters etc can thus be accounted for through the loss of favorable conditions but how by Natural Selection? This puzzles me and yet it seems all the while as if an answer was near by.

The great difficulty seems to lie in this. That Trochiformis begins to show degradational characteristics in the very bed where this variety is most numerous and has the largest individuals. This also occurs among the Ammonites and Nautiloids.

I cannot deny the continual action of Natural and Sexual Selection among animals as I see them in the present fauna of the earths surface and this law must be applicable here. But, how is it possible to apply it when a whole series of species seem to run through a given cycle of changes; a cycle which may be predicted, when it is normal, and be explained by unfavorable conditions, which retard, or interfere with the regular accelerated mode of the inheritance of characteristics, when it is abnormal? I shall work upon these Steinheim shells a portion of this winter and try to provide material for thorough study of the locality.

Hilgendorf has made it possible to go much farther and I only regret that the weather prevents me from going there for several months, and am only fearful that I may be obliged to leave this part of the country before my collections and notes are complete.

You will permit me to thank you for the gratifying notice of my paper on Embryology of Fossil Cephalopods.26 I beg also that, if the length of this letter wearies you, my long silence and the earnest desire I have long entertained of presenting my views to you, may be taken into consideration.

I have written very openly, and frankly stated my difficulties with the hope that you will not find them entirely unworthy of your attention and any advice with regard to my future studies which they may show to be necessary would be most gratefully received by me.

With the highest respect I remain | Your Humble Servt | Alpheus Hyatt

P.O. Address is

Cannstatt bei Stuttgart | No 256 Hoffner Strasse | or simply—Stuttgart | Poste Restante.

CD annotations

4.1 Prof … week. 6.3] crossed pencil
7.11 Thus] blue crayon line in margin
8.20 series thus.] ‘[progression] of embryonic or larval stages.’ blue crayon, line blue crayon to diagram
8.41 I used … several formations. 8.43] ‘—But [is much degraded]blue crayon
10.3 extreme old age] ‘do the old Breed?’ pencil
12.1 Thus … given.] ‘Acceleration of development = origin of development or embryonic character.’ pencil
12.13 exact … series. 12.17] scored blue crayon
12.27 heterogeneous … close. 12.30] scored blue crayon
12.34 forms … age. 12.36] scored blue crayon
12.44 regularity … law. 12.48] scored blue crayon
13.7 How can it] after opening square brackets, pencil; pencil cross in margin
14.14 Doubtless … lake, 14.18] scored blue crayon; ‘Weismann’27 blue crayon
14.32 Thus it may … subsequently. 14.35] scored blue crayon
14.37 part of reverters … Selection? 14.39] scored blue crayon


The letter is dated late in November since CD’s letter of 10 October [1872], to which it is a reply, would probably have taken at least six weeks to reach Hyatt via America.
Letter to Alpheus Hyatt, 10 October [1872].
See letter to Alpheus Hyatt, 10 October [1872] and n. 4. Hyatt refers to Hyatt 1866.
Hyatt refers to Edward Drinker Cope. See letter to Alpheus Hyatt, 10 October [1872] and n. 3.
Louis Agassiz.
Stuttgart (the address from which Hyatt writes) was the capital of the state of Württemberg. (Bad Cannstatt is an outer district of the city.)
The first two pages of Hyatt’s letter are written on two sides of a half sheet of notepaper (a full sheet being a folded piece of paper that gave four sides to write on). The rest of the letter consists of three full sheets and another half sheet, the verso of which is blank. The first half sheet ends at the end of the sixth paragraph (‘until the next week.’).
Albert Oppel was a student of the German palaeontologist Friedrich August Quenstedt. Quenstedt divided each of the three existing divisions of the Jurassic period into six zones in Germany using both palaeontological and geological evidence. Oppel divided the Jurassic formation into thirty-three zones characterised by particular animal species, mostly ammonites, allowing Jurassic deposits to be correlated across different countries. (DSB.)
The Lias is the lower Jurassic; the Oolite the upper (more recent).
In Hyatt 1866, p. 203, Hyatt had described the young of higher species as ‘constantly accelerating their development, and reducing to a more and more embryonic condition or passing entirely over the stages of growth corresponding to the adult periods of preceding or lower species’.
Hyatt probably refers to CD’s ‘rule’ that ‘at whatever period of life a peculiarity first appears, it tends to re-appear in the offspring at a corresponding age, though sometimes earlier’ (Origin 6th ed., p. 10). CD continued, ‘But hereditary diseases and some other facts make me believe that the rule has a wider extension, and that, when there is no apparent reason why a peculiarity should appear at any particular age, yet that it does tend to appear in the offspring at the same period at which it first appeared in the parent’ (ibid.). CD believed that he had misrepresented Hyatt’s and Cope’s views (see letter to Alpheus Hyatt, 10 October [1872] and n. 5).
For Hyatt’s view, see n. 10, above. In Origin, p. 449, CD had maintained, ‘As the embryonic state of each species and group of species partially shows us the structure of their less modified ancient progenitors, we can clearly see why ancient and extinct forms of life should resemble the embryos of their descendants,—our existing species.’ Although a similar passage appears in the sixth edition, it is in the new chapter seven, ‘Miscellaneous objections to the theory of natural selection’, rather than in the chapter on embryology (chapter thirteen in the first edition). CD had written, ‘Hence it is that existing species during the early stages of their development so often resemble ancient and extinct forms belonging to the same class’ (Origin 6th ed., p. 203).
For Cope’s concept of retardation, see Cope 1871, pp. 230–1. Cope had sent an offprint of this paper; CD’s annotated copy is in the Darwin Pamphlet Collection–CUL.
Hyatt 1870. CD’s annotated copy is in the Darwin Pamphlet Collection–CUL.
The Tuberculatusbett and Bucklandibett were adjoining zones or divisions of strata within the lower Lias specified by Oppel in his study of the Jurassic formations of England, France, and south-west Germany (Oppel 1856–8, p. 14).
See Hyatt 1870, pp. 31–2. Hyatt argued that some shells entered the old age or senile period of growth before attaining the fullest adult condition and concluded, ‘They may be said to be arrested in development as far as size is concerned, and retrogressive in development when the reversionary characteristics are considered.’
The Obtususbett was a division immediately above the Tuberculatusbett in Oppel’s system (see n. 15, above). Oppel did not specify a ‘Psilonotubett’ but did refer to the ‘Bett des A. planorbis’ (the lowest division of eight of the lower Lias; see Oppel 1856–8, p. 14). Ammonites planorbis and Psiloceras psilonotum are synonyms of Psiloceras planorbis. Hyatt probably refers to Ammonites collenoti (now Oxynoticeras collenoti), which Oppel identified in the Oxynotusbett, the division above the Obtususbett (ibid., p. 59). No species with the epithet ‘pseudostellare’ has been identified. The Arietitidae is a family of ammonites. Baculites and Orthoceras are two superficially similar but not closely related extinct genera of cephalopods, the former from the late Cretaceous period, the latter from the middle Ordovician. In Hyatt 1866, pp. 197–201, Hyatt discussed what he referred to as the ‘morphological polarity’ of the Ammonoid and Nautiloid series and used the terms ‘Baculite’ and ‘Orthoceratite’ as descriptive terms referring to morphological characteristics of ammonites.
According to Oppel 1856–8, p. 35, the strata associated with Ammonites bucklandi (Bucklandibett) was equivalent to the English ‘blue Lias’. Oppel associated A. brooki (now Caenisites brooki) with the Obtususbett, two levels above the Bucklandibett (ibid., p. 14). Hyatt refers to Alcide Charles Victor Dessalines d’Orbigny.
Oscar Fraas.
See letter to Alpheus Hyatt, 10 October [1872].
In early classificatory systems Ammonites was divided into fourteen families, including Falciferi (for more on early classification of ammonites, see Donovan 1994). The taxon Clyperformae has not been identified.
Franz Hilgendorf had discovered a series of fossil snails in the Steinheim crater in Heidenheim, Germany; these were described in Hilgendorf 1866. Two annotated copies of Hilgendorf 1866 are in the Darwin Pamphlet Collection–CUL; CD cited it in Origin 5th ed., p. 362.
Hyatt probably meant Palustris; P. multiformis was the name Hilgendorf applied to the series of Miocene fossil snails from Steinheim (see letter to August Weismann, 5 April 1872 and nn. 5 and 6). He identified ten stratigraphic zones, each associated with one or more varieties of P. multiformis (Hilgendorf 1866, pp. 478–9).
Hilgendorf 1866 contains a plate that shows the gradations of form of the different varieties of Palustris multiformis and includes a phylogenetic tree.
Palustris multiformis trochiformis was a large variety that occurred in zone six of Hilgendorf’s system. Palustris multiformis revertens was in zone nine and P. multiformis supremus in zone ten (Hilgendorf 1866, p. 479).
Hyatt refers to Hyatt 1872 (see letter to Alpheus Hyatt, 10 October [1872] and n. 2).
See n. 23, above.


No need to apologise for not quoting AH’s paper on acceleration and retardation.

Agassiz introduced AH to ammonites and entrusted collection to him. Has followed developmental history of each species and placed them within geological formations. Found evolutionary history of species recapitulated only to a degree in individual development. Stages frequently skipped. Explains why young of later animals are like adults that preceded them. Retardation entirely idea of Edward Drinker Cope. Sends paper to explain it. Acceleration can explain degraded forms. Often like youthful stage with which series began. Often resemble old age of earlier series. Regularity of these series incompatible with natural selection. How can selection account for degraded final stages or for predictability of development? Franz Hilgendorf’s Paludinae from Steinheim lake show same parallelism in development. May be possible to reconcile this with selection. But Trochiformis begins to show degradation in beds where it is most numerous and has largest individuals, i.e., where selection seems to be favouring it. Will work on Steinheim shells this winter.

Letter details

Letter no.
Hyatt, Alpheus
Darwin, C. R.
Sent from
Source of text
DAR 99: 48–55
Physical description
15pp †

Please cite as

Darwin Correspondence Project, “Letter no. 8655,” accessed on 27 February 2017,