ORIGINAL PAPER

American precursors of evo-devo: ecology, cell lineage,and pastimes unworthy of the Deity

Scott F. Gilbert

Received: 3 September 2007 / Accepted: 10 June 2008 / Published online: 3 July 2008

� Springer-Verlag 2008

Abstract The American precursors of evo-devo have

numerous phenotypes. Fritz Muller, a German emigre liv-

ing in Brazil, was one of the first post-Darwin evolutionary

biologists to look seriously at the roles of larvae in con-

straining and permitting evolutionary change. His book,

Fur Darwin, contains the germs of numerous ideas con-

cerning recapitulation, larval ecology, punctuated

equilibrium, and canalization. William Keith Brooks was

interested in larval ecology and the mechanisms that pro-

moted selectable variation. One of his students, E. B.

Wilson, followed one of Mulller’s paths and brought the

notion of embryonic homologies into the area of develop-

mental biology and animal classification. Frank R. Lillie

took a different page out of Muller and emphasized larval

adaptations.

Introduction

When one discusses ‘‘American’’ precursors to evolutionary

developmental biology, the term has to include both

hemispheres. For naturalists (including Humboldt, Wallace,

Bates, and Darwin), South America, with its rainforests and

teaming littoral zones, was far more important than the

northerly continent. Indeed, it has been argued (Todd 2007)

that Maria Sibylla Merian (1647–1717) was the founder of

ecological developmental biology, having documented in

1679 that caterpillars needed particular food plants on

which to live before metamorphosing into butterflies.

Merian spent much of her active scientific life in Surinam,

as a member of a Pietist mission to this South American

country.

But since I want to look at scientists who followed

Darwin’s Origin of Species, I would not discuss these

earlier naturalists. I also want to look at scientists who were

writing at the time of Darwin, i.e., in the earliest strata of

science influenced by his transformational hypothesis.

Therefore, I will discuss two people in particular, Fritz

Muller and William Keith Brooks. I will end by analyzing

a debate between two younger scientists, Edmund B.

Wilson and Frank R. Lillie, because they were embryolo-

gists who brought an ‘‘American’’, cell lineage, perspective

onto the disputes concerning the evolutionary questions of

homology and adaptation.

Fritz Muller

My first example of a precursor of evolutionary develop-

mental biology in the Americas will be another German

expatriate to South America, Fritz Muller. Whereas Merian

was a religious woman, Muller was a strongly committed

atheist who fled the Prussian scientific establishment to

enter a German agricultural community in Brazil. Both of

these Germans found less restrictive ways of life in

America and were able to pursue science as a woman

(Merian) and as an atheist (Muller). For biographical

details, one should consult Todd (2007) and West (2003).

Johann Friedrich Theodor (Fritz) Muller was born near

Erfurt, Germany, in 1822. In the 1840s, he became a medical

student and an atheist. This latter distinction was the result of

a difficult decision, since he was the son of a Lutheran

minister. Muller became a member of a community that

believed in something like free love, and he refused to take

S. F. Gilbert (&)

Department of Biology, Swarthmore College,

Swarthmore, PA, USA

e-mail: sgilber1@swarthmore.edu

123

Theory Biosci. (2008) 127:291–296

DOI 10.1007/s12064-008-0047-7

his medical oath, because it demanded he swear by ‘‘God and

His sacred Gospel’’ (West 2003; Mallet 2004). As a result of

the conservative clampdown on academic freedom and free

speech after the failed Prussian revolution of 1848, he emi-

grated in 1852 to the German colony of Blumenau in Brazil.

After building some homes and clearing some land, he

became a traveling naturalist for the Brazilian National

Museum. This allowed him to get rid of his despised shoes

and walk barefoot through the Atlantic rainforest.

He became best known for his observations and mathe-

matical analysis of Mullerian mimicry. In 1879, he used

simple algebra and some assumptions about predatory

behavior to show that one unpalatable, warningly-colored

species would benefit from resemblance to another unpala-

table species by a factor equal to the square of the inverse

ratio of the species’ relative abundances. This was the first

example of frequency-dependent selection, as well as the

first mathematical treatment of mutualism. Thus, Muller

pioneered mathematical evolutionary ecology (see Mallet

2007).

Muller also became known for his book, Fur Darwin

(1864; translated into English as Facts and Arguments for

Darwin in 1869). This small (less than 150 pages) volume

championed a program wherein the goal of embryology

was to reconstruct phylogenies. It combined natural

selection with embryology to demonstrate that ‘‘Darwin’s

theory furnishes the key of intelligibility for the develop-

mental history of crustaceans, as for so many facts

inexplicable without it’’. He compared embryonic stages

between species, believing that ‘‘above all things, a thor-

ough knowledge of development’’ is critical for evolution

to work in explaining phylogenies (Muller 1869, p. 4).

Muller wrote engagingly of adaptations of those

organisms he saw in the tidal basins of Brazil. First, the

crustaceans were a perfect place to see natural selection in

action. Indeed, Muller says that some sort of transforma-

tionism had been assumed among the crustacean

researchers (1869, p. 3):

Among the parasitic Crustacea, especially, everybody

has long been accustomed to speak, in a manner

scarcely admitting of a figurative meaning, of their

arrest of development by parasitism, as if the trans-

formation of species were a matter of course. It would

certainly never appear to anyone to be a pastime

worthy of the Deity, to amuse himself with the con-

trivance of these marvellous cripplings, and so they

were supposed to have fallen by their own fault, like

Adam, from their previous state of perfection.

His arguments against Agassiz’s model of special cre-

ation is a point-by-point scientific rebuttal (Muller 1869,

p. 28); and he often expresses his views in a colloquial

manner (as witty as Thomas Huxley, but without Huxley’s

smiling malice). On p. 27, he introduces the reader to two

species of Amphipods-lusty crustaceans both of whom live

in the intertidal zone and are observed in copulation more

often than not. (One species is called Melita insatiabilis;

the other, named after the slatternly Roman empress,

Melita messalina) (Muller 1869, p. 27):

The two species in which I am acquainted with this

structure are amongst the most salacious animals of

their order, even females which are laden with eggs in

all stages of development, not unfrequently have their

males upon their backs.

However, in one species the female has a genital clasper

allowing her to keep the male bonded to her during the

tidal fluxes, while the other species does not. So Muller

continues (1869, p. 29):

Its presence only in these few Amphipoda will have

to be regarded not as the work of far-seeing wisdom,

but as that of a favourable chance made use of by

Natural Selection. Under the latter supposition its

isolated occurrence is intelligible, whilst we cannot

perceive why the Creator blessed just these few

species with an apparatus which he found to be quite

compatible with the ‘‘general plan of structure’’ of the

Amphipoda, and yet denied it to others which live

under the same external conditions, and equal them

even in their extraordinary salacity.

Expanding Darwin’s idea that ‘‘community of embryo-

nic structure reveals community of descent’’, Muller wrote

that homologous larval structures indicated shared ances-

try. Thus, he proclaimed the Nauplius larva to be the

common source of all crustaceans, and he declared that its

basic structure was that of the crustacean ancestor. Having

such a larva became the criterion for membership in the

crustaceans, and Muller demonstrated that several parasitic

species thought to be worms were actually crustaceans by

virtue of their going through a Nauplius stage (see Tauber

and Chernyak 1991). Muller closes his book with the

suggestion that if we look for the common ancestor of

crabs and insects, we should expect to find ‘‘a zoea which

raised itself into life on land’’ (1869, p. 141).

Muller (1869, p. 118) also argued for the efficacy of

natural selection both in the adults and in larval stages:

For it is perfectly evident that the struggle for exis-

tence and natural selection combined with this, must

act in the same way, in change and development,

upon larvae which have to provide for themselves, as

upon adult animals.

These larval adaptations create a ‘‘falsification’’ of the

record preserved in the developmental history, because

adults and their larvae both evolve adaptations to survive in

292 Theory Biosci. (2008) 127:291–296

123

their respective environments. Moreover, because of larval

adaptations, scientists should not be surprised when perfect

reflections of phylogeny are not seen in the embryonic

record of extant organisms.

Muller’s little book is a mine of incredible ore. In it, one

sees the anlagen of our current hypotheses of canalization,

developmental constraints, and even punctuated equili-

brium. He proposed a type of canalization and

heterochrony wherein plastic traits that seen later in life

eventually become an inherited property (1869, p. 114):

Thus as the law of inheritance is by no means

strict, as it gives room for individual variations with

regard to the form of the parents, this is also the

case with the succession in time of the develop-

mental processes […]. A precocious appearance of

peculiarities acquired at a later period will generally

be advantageous, and their retarded appearance

disadvantageous; the former when it appears acci-

dentally, will be preserved by natural selection. It is

the same with every change which gives to the

larval stages […] a more straightforward direction,

simplifies and abridges the process of development,

and forces it back into an earlier period of life, and

finally into the life of the egg.

Developmental constraints are discussed in relation to

why some parts of the body are more amenable to change

than others. Respiratory apparatuses might change, he

notes, to make an organism more fit in a particular

environment; but ‘‘the primitive form of the heart was

inherited unchanged, because any variations which might

make their appearance were rather injurious than advan-

tageous, and disappeared again immediately’’ (Muller

1869, p. 44).

Punctuated equilibrium is also assumed: ‘‘The historical

development of a species can hardly have taken place in a

uniform flow; periods of rest have alternated with periods

of rapid progress’’ (Muller 1869, p. 115).

Muller’s brand of recapitulationism is very complex.

Indeed, the intereractions between Ernst Haeckel and Fritz

Muller are rather complicated. According to West (2003),

it was Haeckel who probably alerted Darwin to Muller’s

German book in October, 1864; and it was also Haeckel

who probably adopted Muller’s description of recapitula-

tion for his own 1866 book. Breidbach (2006) claims that

Haeckel’s theoretical formulation of recapitulation was not

able to determine which traits were phylogenically pre-

served (palingenetic) and which were adaptations to the

specific environmental circumstances of the embryo

(cenogenetic). Muller’s analysis, however, provided a

paradigmatic example (i.e., the crustaceans) of how com-

parative embryology could be made into an evolutionary

morphology and how certain anatomical states (such as the

nauplius and zoea larva) could be considered preserved and

representatitive of the developmental trajectory.

When Muller sent Alexander Agassiz (Louis’ son) his

book in 1865, Agassiz replied (in West 2003, p. 137):

I have read very carefully your Fur Darwin and I was

much pleased to see the first beginning of an attempt

to test ‘Darwin’ by facts especially by facts applied to

Embryology. It has always appeared to be a great

oversight in the supporters of Darwin not to take hold

of Embryology (where they would find) much more

substantial evidence than the conclusions thus far

drawn from different breeds under the influence of

man.

Facts for Darwin is a quick and worthwhile read for any

evolutionary developmental biologist. If nothing else, it

gives one an appreciation for the man whom Darwin called

‘‘the prince of observers’’.

William Keith Brooks

Another American biologist who felt that developmental

biology was critical for evolution was the Johns Hopkins

morphologist William Keith Brooks. Brooks is known,

when he is known at all, as the thesis advisor for Thomas

Hunt Morgan, E. B. Wilson, Edwin G. Conklin, Ross

Granville Harrison, and as the person who got William

Bateson interested in Balanoglossus (see Benson 1987;

Maienschein 1987, 1991). Needless to say, the embryonic

question of autonomous versus induced determination

comes to mind. Was Brooks an incredibly gifted teacher, or

was he just a lucky professor who managed to attract

outstanding graduate students? As in embryology, probably

a mixture of the two presided.

During his lifetime, however, Brooks was known for his

scientific study of the Virginia oyster, a pioneering con-

servation biology study that showed the importance of

ecological factors for larval development and demonstrated

the variation between closely related species. The impor-

tance of substrates for larval settlement and metamorphosis

was first demonstrated in 1880, when Brooks, an embryo-

logist at Johns Hopkins University, was asked to help the

ailing oyster industry of Chesapeake Bay (Brooks 1880;

see Keiner 1998). For decades, oysters had been dredged

from the bay, and there had always been a new crop to take

their place. But recently, each year brought fewer oysters.

What was responsible for the decline? Experimenting with

larval oysters, Brooks discovered that the American oyster

(unlike its better-studied European cousin) needed a hard

substrate on which to metamorphose. For years, oystermen

had thrown the shells back into the sea, but with the advent

of suburban sidewalks, the oystermen were selling the

Theory Biosci. (2008) 127:291–296 293

123

shells to the cement factories. Brooks’ solution: throw the

shells back into the bay. The oyster population responded,

and the Baltimore wharves still sell their descendants. So

eco-devo in America was intimately tied to sustainability

and resource management from the outset.

Brooks was fascinated by the notion of variation and its

propagation, and his big book in this area was The Law of

Heredity. A Study of the Cause of Variation and the Origin

of Living Organisms (1883). Dedicated to the memory of

Charles Darwin, it claims that what Darwin needed was a

theory of heredity that could close the two holes in his

theory: first, how can change occur such that the organism

can develop according to a new path?

How are the various organs of a highly complicated

organism, or the various structures which enter into

the formation of a complicated organ, kept in har-

monious adjustment to each other by the selection of

variations which are, in Darwin’s sense, fortuitous?

(Brooks 1883, p. 281)

.

Second, how can such a rare alteration be propagated,

when only one member of the species has such a changed

structure?

Here, he quotes Darwin as to the inability of natural

selection to cause variation. Most developmental biologists

who have read Darwin at all probably have read only The

Origin of Species. We have to remember that Origin was

only Darwin’s first book on evolution. Darwin realized that

selection could not act upon traits that had not yet

appeared, noting that ‘‘characters may have originated from

quite secondary sources, independently from natural

selection’’ (Darwin 1859, p. 196). He continued this line of

reasoning in his book on variation and domestication

(Darwin 1883, p. 282), wherein he admits that:

The external conditions of life are quite insignificant,

in relationship to any particular variation, in com-

parison with the organization and constitution of the

being which varies. We are thus driven to conclude

that in most cases the conditions of life play a sub-

ordinate part in causing any particular modification.

Brooks reminds us that Darwin admitted that natural

selection could not cause variation, and that a theory of

variation was needed to supplement the theory of natural

selection. Indeed, Brooks called one of his book’s chapters:

‘‘The theory of heredity considered as supplementary to the

theory of natural selection’’. Brooks, for whom Darwin’s

book on variation was current reading, quotes that passage

and others to show the need for such a theory of heredity.

For him as for other late nineteenth century biologists,

‘‘heredity’’ is a term encompassing the fields of both

development and genetics.

What was Brooks’ model? It is based on adaptive

plasticity and its propagation to the next generation through

gemmules made by the responding cells. On p. 293, we

read:

According to our theory of heredity, when an

organism, placed under new conditions, becomes

modified to meet the change in its environment, the

existence of the internal change is caused by the

external change, while its precise character is deter-

mined by other factors, chiefly by the hereditary

characteristics of the corresponding part, in both

parents. As long as the harmony which has been

gradually established, by natural selection, between

any particular cell and its conditions of life, remains

undisturbed, this cell will continue to perform its

function as a part of the body, and will have little

tendency to give rise to gemmules […]. These

gemmules, when transmitted to the egg, by impreg-

nation, will, by sexual union with the corresponding

parts of the egg, cause variation in the homologous

cells’ of the offspring, and will thus produce a con-

genital hereditary change at the very time when, and

in the very part where, such change is needed.

(Brooks 1883, p. 293)

Here, then we see a model wherein the phenotype leads

the genotype and the developmental plasticity seen in the

parent generation can be transmitted to the filial generation.

The developmental orderliness is maintained because these

phenotypic changes were originally physiological ones that

the organism could make. The propagation of these vari-

ants was assured because the physiological change

occurred in many of the individuals simultaneously. What

Brooks seems to be proposing is an epigenetic cascade. The

parts of the embryo would have to interact harmoniously.

Moreover, according to Brooks’ view, the changes made

are not merely ‘‘fortuitous’’. The environment biased which

changes are made and which changes can be propagated.

Gemmules, those particles thought to be given off by adult

tissues and which are absorbed by the eggs as hereditary

determinants, have long passed out of science. But we can

see what he was striving for—a mechanism by which

physiological traits can be converted into hereditary traits,

the beginnings of genetic assimilation and phenotypic

accommodation. Indeed, such speculations were in the air

and included Spalding’s 1873 speculations as well as the

triad of Baldwin, Lloyd Morgan, and Osborn, each in 1896.

Frank R. Lillie and Edmund B. Wilson

The next pair of American contributors to the tributaries of

evo-devo are Frank R. Lillie and Edmund Beecher Wilson.

294 Theory Biosci. (2008) 127:291–296

123

As the nineteenth century gave rise to the twentieth, these

were two of the most important biologists in that scientific

backwater called the United States of America. In 1898,

these two eminent embryologists gave cell lineage lectures

at the Marine Biological Laboratory at Woods Hole,

Massachusetts, and these well-publicized lectures served to

emphasize the two ways in which embryology was being

used to support evolutionary biology.

The first lecture, presented by Brooks’ student, E. B.

Wilson, was a landmark in the use of embryonic homo-

logies to establish phylogenetic relationships. Entitled

‘‘Cell Lineage and Ancestral Reminiscence’’ Wilson

(1898), had observed the spiral cleavage patterns of flat-

worms, molluscs, and annelids, and he had discovered that

in each case, the same organs came from the same groups

of cells. This was especially true in the manner that these

animals formed their mesoderm from teloblasts, large cells

set aside early in cleavage. From the two subsequent

teloblasts arose an orderly succession of smaller cells

arranged in a chain-like manner. These mesodermal bands

characterized the annelids (Lumbricus and Nereis), the

mollusks, and the polyclad flatworms. For him this meant

that these phyla all had a common ancestor. Here, Wilson

was using developmental process rather than a develop-

mental structure as the basis for homology. It wasn’t that

the mesoderm was homologous; it was the way of forming

the mesoderm that was homologous. Indeed, modern

research using DNA sequences has confirmed Wilson’s

conclusion and placed these three phyla together.

The other lecturer was F. R. Lillie, who had also done

his research on the development of molluscan embryos and

on their cell lineages. He stressed the modifications, not the

similarities, of cleavage. He presented his research on

Unio, a mussel whose cleavage pattern is altered to pro-

duce the ‘‘bear-trap’’ larva that enables it to survive in

flowing streams.

Streams create a problem for the dispersal of larvae:

because the adults are sedentary, free-swimming larvae

would always be carried downstream by the current. These

clams, however, had adapted to this environment via

modifications of their development. The first is an alter-

ation in embryonic cleavage. In typical molluscan

cleavage, either all the macromeres are equal in size or the

2D blastomere is the largest cell in the 16-cell embryo.

However, cell division in Unio is such that the 2d blasto-

mere gets the largest amount of cytoplasm. This cell

divides to produce most of the larval structures, including a

gland capable of producing a large shell. The resulting

larvae (called glochidia) resemble tiny bear traps; they

have sensitive hairs that cause the valves of the shell to

snap shut when they are touched by the gills or fins of a

wandering fish. The larvae attach themselves to a fish and

‘‘hitchhike’’ with it until they are ready to drop off and

metamorphose into adult clams. In this manner, they can

spread upstream. Lillie (1898) argued that ‘‘modern’’

evolutionary studies would do better to concentrate on

changes in embryonic development that allowed for sur-

vival in particular environments than to focus on ancestral

homologies that united animals into lines of descent.

Conclusions

Thus in the Americas of 1898, some major approaches

relating evolution and development were clearly defined:

one was to find underlying unities that link disparate

groups of animals. The second sought to detect those dif-

ferences in development that enable species to adapt to

particular environments. Darwin thought these two

approaches to be temporally distinguished—that is, that

one would find underlying unities in the earliest stages of

development, while the later stages would diverge to allow

specific adaptations (see Ospovat 1981). However, Wilson

and Lillie were both discussing the same stage of

embryogenesis, cleavage. These remain three of the major

research programs in evolutionary developmental biology.

The third path consisted of looking at developmental

plasticity as a mechanism that may allow the spread of a

phenotype widely through a population.

References

Baldwin JM (1896) A new factor in evolution. Am Nat 441–451:536–

553

Benson KR (1987) H. Newell Martin, W. K. Brooks, and the

reformation of American biology. Am Zool 27:759–771

Breidbach O (2006) The conceptual framework of evolutionary

morphology in the studies of Ernst Haeckel and Fritz Muller.

Theory Biosci 124:265–280

Brooks WK (1880) Development of the American Oyster. Report of

the commission of fisheries of Maryland. Inglehart, Annapolis

Brooks WK (1883) The law of heredity. A study of the cause of

variation and the origin of living organisms, 2nd edn (revised).

John Murphy, Baltimore

Darwin C (1859) On the origin of species by means of natural

selection. John Murray, London

Darwin CD (1883) The variation of animals and plants under

domestication. Appleton, New York

Keiner C (1998) W. K. Brooks and the oyster question: science,

politics, and resource management in Maryland, 1880–1930.

Journal of the History of Biology 31:383–424

Lillie FR (1898) Adaptation in cleavage. Biological lectures from the

marine biological laboratory. Woods Hole, pp 43–67

Lloyd Morgan C (1896) On modification and variation. Science

4:733–740

Maienschein J (1987) H. N. Martin and W. K. Brooks: exemplars for

American biology? Am Zool 27:773–783

Maienschein J (1991) Transforming rraditions in American biology,

1880–1915. Johns Hopkins University Press, Baltimore

Mallet J (2004) Review of Fritz Muller: a naturalist in Brazil. Quart

Rev Biol 79:196

Theory Biosci. (2008) 127:291–296 295

123

Mallet J (2007) Muller. http://www.ucl.ac.uk/taxome/jim/Mim/

Muller.html. Accessed 26 April 2008

Merian MS (1679) Der Raupen wunderbare Verwandlung und sonder-

bare Blumennahrung [The Caterpillar, Marvelous Transformation

and Strange Floral Food]. Joannem Oosterwyk, Nuremberg

Muller F (1869) Facts and arguments for Darwin. John Murray,

London

Muller F (1879) Ituna and Thyridia; a remarkable case of mimicry in

butterflies (transl. by Ralph Meldola from the original German

article in Kosmos, May 1879, 100). Transactions of the

Entomological Society of London, 1879. pp xx–xxix

Osborn HF (1896) Ontogenic and phylogenic variation. Science

4:786–789

Ospovat D (1981) The development of Darwin’s theory. Cambridge

University Press, Cambridge

Spalding DA (1873) Instinct with original observations on young

animals. Macmillan’s Mag 27:282–293

Tauber AI, Chernyak L (1991) Metchnikoff and the origins of

Immunology: from metaphor to theory. Oxford University Press,

New York

Todd K (2007) Chrysalis: Maria Sibylla Merian and the secrets of

metamorphosis. Harcourt, New York

West DA (2003) Fritz Muller: a naturalist in Brazil. Pocohantas Press,

Blacksburg

Wilson EB (1898) Cell lineage and ancestral reminiscence. Biological

Lectures from the Marine Biological Laboratories. Woods Hole,

pp 21–42

296 Theory Biosci. (2008) 127:291–296

123