skip to content

Darwin Correspondence Project

From Alexander Agassiz   [before 1 June 1871]1

Museum of Comparative Zoology, | At Harvard College, | Cambridge, Mass.

My dear Mr. Darwin

There are some points in the sexual differences of Viviparous fishes from California (Embiotocoids) which are suggested to me by reading your chapters on sexual selection   The males and females can be distinguished at a glance by their anal fins. In some genera the shape of the fins even is different in the two sexes, viz in Amphyticus, Metrogaster, Holcenotus, & Hysterocarpus—while in the other genera of the family Micrometrus, Hyperprosopon, Phanerodon, Rhachochilus, Embiotoca, the fins (anal) in the two sexes are similar in shape, but we find invariably in the males a fleshy accumulation near the base of the anal while the anal fins of the females are perfectly smooth.2 In the genera first mentioned this fleshy accumulation is placed in the central part of the fin, what the use of this is I cannot say. It is found in small specimens already and does not appear simply at breeding season. The accompanying sketch shows this better than any description I can give. How the impregnation takes place in these fishes I do not know and no fisherman in San Francisco ever could give me any information on the subject3   There must evidently from the nature of the case be copulation yet the males have no copulative organ and no means of retaining the sperm near the sexual openings except this fleshy accumulation on the anal fin which may perhaps serve some such purpose in the process of fecundating the females, though how the sperm reaches the eggs I cannot imagine as the eggs when immature are far up in the ovarian bag, at a considerable distance from the sexual opening, to preserve the young and prevent the disappearance of the species? the small number of young rarely over 18—are retained in the pouch of the female and leave it only as full fledged, scaly fishes, frequently 12 as long as the mother. There is apparently but one breeding season in the year & the species would soon die out, were the 18 or less eggs which come to maturity left to float about at the mercy of the winds & waves and take their chance of becoming impregnated & afterwards to grow to maturity. The amount of milt of one of the males is quite considerable entirely out of proportion to the number of eggs to be fecundated & as far as my experience with the fishermen & in the San Francisco market went there is not a larger number of females than of males.

One other point suggested by some remarks made by Mivart in his Genesis of Species.4 He seems to look upon the pedicellaria as such inexplicable organs   Müller and I have both conclusively shown that pedicellariae are modified spines.5 Their development as far as I have seen it in Starfishes and in Seaurchins leaves no doubt on this point. & their function in the Seaurchins at least in those where they are most highly developed & exist in greatest numbers is that of Scavengers   they keep the ambulacral suckers free from any impurities which may become entangled in the ambulacral spaces, and have found their way there either as excrements from the apex or have been brought there by currents. & it is a very amusing sight to see the rapidity with which these small forceps will work and clear away any rubbish which comes in their way & throw it off the Test of the Seaurchin, passing it from one forcep to the other. It is as early a sign of intelligence as I have noticed in lower animals except the case of maternal sollicitude displayed by a Starfish for its eggs.6

We are progressing finely with the addition to our Museum, father I am happy to say is gaining daily in strength.7 The doctors advise him to absent himself for some time from Cambridge to ensure his total recovery and he talks of making a trip to California via Cape Horn! in one of the Government Vessels, which is to be supplied with means for deep sea dredging on the way.8

With kindest remembrances for Mrs Darwin from Mrs Agassiz to yourself & the members of your household | I am always | very truly yours | Alex. Agassiz


Pedicellariae with tripartite pincers are found in all Seaurchins   In Starfishes we find not only pedicellariae with tripartite pincers (Luidia) but also pedicellariae in which one of the forceps is reduced to a support of the other two. (Asteracanthion); while in other genera there are the so called valvular pedicellariae (Oreaster, Astrogonium) in which the two pincers alone exist.10 We have all possible gradations between what are called simple granules+ in such genera as Oreaster to the valvular pedicellariae of the same genus to the tripartite pincers with one rudimentary member of Asteracanthion, to the genuine tripartite pincers of Luidia ressembling the tripartite pedicellariae of Echinidae but unlike them not provided with a moveable stem; but merely perched upon a support. Some of the pedicellariae of Asteacanthion are already provided with a short muscular stem but not a calcareous one   In regular Echini we find the tripartite peddicellariae supported upon a stem with a calcareus rod inside and capable of motion very much like that of the ordinary spines11   This calcareus rod+ is articulated generally upon a miliary tubercle (a simple granule) analogous to that of the spines, but in one of the genera dredged by Mr Pourtales, Podocidaris I found pedicellariae articulated upon a tubercle provided with a mamelon a scuticular circle in all respects similar to that of the ordinary spine of the Seaurchin.12 In the Spatangoids13 we find what are called fascioles bands of minute granules supporting embryonic spines articulated upon these bands limited to certain regions of the body performing the functions of regular pedicellariae of keeping the dirt from the ambulacral petals, and at same time they are provided with tripartite pedicellariae more complicated & more highly organized than in any other Echinoderms, scattered irregularly over the test. In our common Sea urchin the pedicellariae (Toxopneustes drobachiensis)14 are used to remove the dirt from the test & keep it from getting among the suckers. We never find strictly branching spines among Echini we have nothing more than very long spines sometimes equalling the diam of the main spine though in very young Seaurchins it would often be difficult to say they were not for their fantastic shape.

But we do have in Starfishes in the genus Luidia the very combinations needed to shew that the pedicellariae are only a modified branching spines   we have a central prolongation of the fixed spine and around this are placed the three pincers of the body sometimes even with a second set of them at base of the spine. The following rough sketches will shew you the whole argument better than my description.


Asteracanthion with a short flexible stem (muscular




immoveable base pincers can simply open & shut


Same structure as previous one in Luidia. but the spine nature is more apparent 3 branches at base of immovable support of tripartite pincers.


Going backward a little you will see that we find all possible variation between the pedicellariae of Starfishes and hooks, of Ophiurans.15 and in same way all possible gradations between the pedicellariae of Seaurchins and the anchors and plates on the dermal portion of Holothurians,16 & all these calcareus deposits are modified spines highly specialised in Echinidae as prehensile organs. Their early development is the same of all these bodies as accumulations of lime cells rising above the general level, covered by a layer of epithelial, and this goes on till a fixed spine is formed. as is still found in some genera of Echinidae (both fixed & moveable exist on it) viz. Podocidaris and Glyptocidaris (a fossil of Jurassic period)17 or else it develops a joint & forms the usual spine of the Seaurchin or it is developed into the fixed pedicellariae of Starfishes or the moveable ones of Echini with jointed stems.


1 piece of test of seaurchin with very young spines fixed.

2 older. spine still fixed   tubercle appears .t.

3 connection between tubercle of spine very slight. older than 3.

4 full sized spine with bossed ring striated longitudinally   in all former figures spines made up of larger or smaller cells more or less packed closely according to age.

5 is a fixed spine like 2 in structure found in an adult Seaurchin Podocidaris from deep water off Florida.

Spines like 3. with an imperfect articulation are characteristic of the spines found on the scaly buccal membrane of Cidaridae, something similar also in the old Silurian Palechinidae.18

outline represents epithelial layer.


1. 2. 3. different stages of growth of spines of pedicellariae & branch of spine of our common Starfish Asteracanthion berylinus19

s   spine

p   pedicellariae

b   branch of spine


Embryonic spine of the fascioles of Spatangoids

+This point I have never pressed sufficiently but am convinced from the similar mode development of granules, pedicellariae, spines of the correctness of the statement in Starfishes and Seaurchins—

+In all the Cidaridae this rod is less moveable still recalls somewhat the structure of pedicellariae of Starfishes

CD annotations

2.1 One … the other. 2.11] crossed pencil; ‘Pedicellariae’ added ink
2.4 Starfishes] ‘Starfish’ added pencil
2.4 Seaurchins] ‘Sea-urchins’ added pencil
2.11 Test] ‘test’ added pencil
2.12 It … eggs. 2.13] double scored blue crayon; ‘Descent of Man’ added blue crayon
3.1 We … way. 3.5] crossed pencil
Top of letter: ‘I do not think the case worth notice, as [1 word illeg, 1 word del illeg] primary sexual difference’ pencil
Top of first page, verso: ‘Fish’ added blue crayon
1.1 In Starfishes … Astrogonium) 1.5] scored blue crayon
1.9 tripartite pedicellariae] underl blue crayon
1.11 short muscular stem] underl red crayon
1.18 what are called fascioles … supporting 1.19] underl blue crayon
1.19 embryonic … petals, 1.21] scored red crayon
1.20 performing the functions] underl red crayon
1.25 We never … shape. 1.28] double scored blue crayon
2.1 But] after opening square bracket red crayon
2.1 But … spines 2.2] double scored blue crayon
3.1 Going … organs. 3.5] scored red crayon


The date is established by the relationship between this letter and the letter to Alexander Agassiz, 1 June [1871].
The fish genera Agassiz mentions are in the family Embiotocidae (surfperches). Amphyticus: presumably Agassiz refers to the genus Amphistichus. Metrogaster is now Cymatogaster while Holchonotus (Agassiz’s ‘Holcenotus’) is now an unaccepted genus, whose species are placed within the genera Embiotoca, Amphistichus, and Micrometrus. The Embiotocidae were first described by Agassiz’s father, Louis Agassiz (J. L. R. Agassiz 1853). On the modifications of the anal fin related to reproduction, see Tarp 1952, pp. 7–10.
Observation of copulation in surfperches was not reported until 1917 (Hubbs 1917); for more on courtship and mating behaviour see Rechnitzer and Limbaugh 1952.
Agassiz refers to St George Mivart and Mivart 1871a.
See Mivart 1871a, pp. 44–5; the section is annotated in CD’s copy in the Darwin Library–CUL (see Marginalia 1: 584–8). For Agassiz’s and Johannes Müller’s observations on the formation of the pedicellariae, see A. Agassiz 1864 and J. Müller 1850–1.
In Origin 6th ed., pp. 191–3, CD added Agassiz’s remarks on the function and development of the pedicellariae in Echinodermata in response to Mivart’s critique.
Agassiz refers to the extension of the Museum of Comparative Zoology at Harvard. The addition was completed in 1871 (Lurie 1960, p. 371). In September 1869, Louis Agassiz had suffered a cerebral haemorrhage that left him paralysed and unable to speak, but he gradually made a full recovery (Lurie 1960, pp. 368–9).
Louis Agassiz travelled on the Hassler, an experimental steamship of the United States Coast Survey, leaving from New York in December 1871, passing through the Straits of Magellan, and arriving in San Francisco in the autumn of 1872 (Lurie 1960, pp. 371–7).
The drawings in the enclosure are reproduced at their original size except fo the drawing of Cidaris, Diadema, Echinus, and Podocidaris, which is reproduced at 60 per cent of its original size, and the series of five sketches of seaurchin spines, reproduced at 80 per cent of their original size.
Asteracanthion is now an unaccepted synonym of Asterias and its species have been placed in several genera within the family Asteriidae; Astrogonium is now an unaccepted synonym of Goniaster, with species placed in several genera within the family Goniasteridae (see the World Register of Marine Species at for taxon details).
The calcareous plates forming the skeleton of an echinoderm are referred to as ossicles. The spines, spinelets, pedicellariae, and granules are referred to as accessory ossicles and they vary in both morphology and function in different genera and species.
Agassiz refers to Louis François de Pourtalès and to the sea urchin genus Podocidaris. In A. Agassiz 1873, p. 402, Agassiz described and illustrated the spines of the specimen of Podocidaris collected by Pourtalès.
The reference is to the order Spatangoida (heart urchins).
Toxopneustes droebachiensis is now Strongylocentrotus droebachiensis, the northern or green sea urchin.
Agassiz refers to the class Ophiuroidea: basket stars, brittle stars, and snake stars.
The reference is to the class Holothuroidea, sea cucumbers.
The genus Glyptocidaris (family Phymosomatidae), the fossil existence of which dates from the early Jurassic period, has only one extant species.
The family Cidaridae are pencil urchins; fossils of the Silurian species Palaeodiscus ferox (family Palaechinidae) had been discovered in Herefordshire by John William Salter; Salter 1857, p. 332.
Asteracanthion berylinus is now Asterias forbesi.


Agassiz, Alexander. 1864. Embryology of the starfish. From J. L. R. Agassiz’s Contributions to the Natural History of the United States vol. 5, published 1877. Cambridge, Mass.: n.p.

Agassiz, Alexander. 1873. The homologies of pedicellariae. American Naturalist 7: 398–406.

Agassiz, Louis. 1853. Extraordinary fishes from California, constituting a new family. American Journal of Science and Arts 2d ser. 16: 380–90.

Hubbs, Carl L. 1917. The breeding habits of the viviparus perch, Cymatogaster. Copeia (1917): 72–4.

Lurie, Edward. 1960. Louis Agassiz: a life in science. Chicago: University of Chicago Press.

Marginalia: Charles Darwin’s marginalia. Edited by Mario A. Di Gregorio with the assistance of Nicholas W. Gill. Vol. 1. New York and London: Garland Publishing. 1990.

Origin 6th ed.: The origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. 6th edition, with additions and corrections. By Charles Darwin. London: John Murray. 1872.

Salter, John William. 1857. On some new Palæozoic star-fishes. Annals and Magazine of Natural History 2d ser. 20: 321–334.

Tarp, Fred Harald. 1952. Fish Bulletin No. 88. A Revision of the Family Embiotocidae (The Surfperches). UC San Diego: Scripps Institution of Oceanography Library.


Instances of sexual differences in viviparous fishes, suggested by reading chapters on sexual selection [in Descent] and by Mivart’s Genesis of species.

Notes on echinoderms.

Letter details

Letter no.
Alexander Agassiz
Charles Robert Darwin
Sent from
Museum of Comparative Zoology, Harvard
Source of text
DAR 69: A43–6 DAR 89: 29–31
Physical description
ALS 8pp † sketches

Please cite as

Darwin Correspondence Project, “Letter no. 7415,” accessed on 15 April 2024,

Also published in The Correspondence of Charles Darwin, vol. 19