The Origin of Speciess09

Chapter VII. Miscellaneous Objections To The Theory Of Natural Selection

Charles Darwin · 1872. Read the original ↗

Thesis

Darwin answers a series of objections, chiefly St. George Mivart's. He argues that natural selection, aided by variation, the laws of growth and the inherited effects of use, can account for the incipient stages of useful structures through graded, serviceable intermediates, and that abrupt origin of species is improbable.

Darwin makes a strong case that graded, plausibly useful intermediates exist and that the 'no incipient stages' objection is not decisive, and one prediction (climbing plants) is well confirmed. But most central claims are inferences about unobserved histories, and several rest on the survival of structures as evidence of advantage or on inherited effects of use that were assumed. The agency language is mostly shorthand for a stated mechanism, but in the progress passage it carries the argument.

CInference

Claims, graded

  1. C1InferredC

    The incipient stages of complex useful structures (baleen, pedicellariae, avicularia, orchid pollinia, crustacean pincers, mammary glands) can be explained because graded series of forms exist, each step plausibly useful, so selection could preserve and improve them step by step.

    “I wish only to show that this is not incredible, and that the immense plates of baleen in the Greenland whale might have been developed from such lamellæ by finely graduated steps, each of service to its possessor.”

    Evidence offered Comparative series among living forms (goose to duck to shoveller beak, spine to pedicellaria, simple to elaborate pollinium, crustacean chelae) and specialists' reports of homology. The usefulness of each step is argued from function, not measured.

    1. 1Observation//The graded series among living species are observed, but the ancestral sequence and the usefulness of each step are not.
    2. 2Hypothesis//The claim is specific enough to fail, since a structure with no plausible useful intermediate would count against it, but many steps are stated only as possibilities.
    3. 3Prediction/Darwin predicts that intermediates should exist, but most are fitted to cases already known, and he disclaims the baleen scenario as a literal claim.
    4. 4Experiment—No controlled test of the fitness of any intermediate stage is offered.
    5. 5Replication/Many naturalists' observations of related structures agree, but none tests the function of the intermediates.
    6. 6Falsifiability/Mivart's rivals (sudden change, internal tendency) are addressed, but the argument could absorb almost any series as support.

    What would move it up Fossil or developmental sequences linking the ancestral and derived structures, plus measurements showing that the intermediate forms really conferred a survival or reproductive advantage.

  2. C2InferredC

    Slight proportional variation in a hoofed ancestor, together with differential survival of the longest-reaching individuals in dearths, would suffice to produce the giraffe. As with human unconscious selection, nature needs no foresight or isolation of pairs, because it preserves the superior individuals while they intercross freely.

    “natural selection will preserve and thus separate all the superior individuals, allowing them freely to intercross, and will destroy all the inferior individuals.”

    Evidence offered Natural-history measurements showing that parts differ slightly in relative length, the giraffe's current browsing advantage, the Niata cattle as an example of a small structural difference affecting survival in droughts, and the race-horse and greyhound as artificial analogues.

    1. 1Observation//Individual variation in proportions and the survival effect of small structural differences (Niata cattle) are observed, but not the giraffe's lineage.
    2. 2Hypothesis//The scenario is concrete (survival during dearths by height of reach), though phrased with 'will often have been' and 'would generally have survived'.
    3. 3Prediction/No specific prediction is made that could fail, and the 'why not other species?' replies are given afterwards.
    4. 4Experiment/Domestic breeding is an uncontrolled-for-nature analogue in which humans, not nature, set the criterion.
    5. 5Replication/Breeding results are widely repeated, but the natural process in giraffes is not independently tested.
    6. 6Falsifiability/Mivart's food-supply and 'why only the giraffe' objections are weighed, but any non-occurrence is explained by absent conditions.

    What would move it up Survival and fecundity data by neck or limb length in living browsers during dearths, heritability estimates for those traits, and a dated fossil series of giraffe ancestors.

  3. C3ObservedB

    Climbing by twining, leaf-clasping and tendrils could arise by selection from slight revolving and touch-sensitive movements, and these movements should therefore be found in non-climbing plants.

    “Hence, I was led to predict that some slight tendency to a movement of this kind would be found to be far from uncommon with plants which did not climb”

    Evidence offered The prediction preceded the discoveries: Maurandia peduncles, Müller's Alisma and Linum, Morren's and Darwin's Oxalis, and Hofmeister's report that young shoots and leaves move after being shaken.

    1. 1Observation///The rudimentary movements were directly seen and reported by several observers.
    2. 2Hypothesis///The claim says exactly what should be found in non-climbers, and it could have failed.
    3. 3Prediction///Darwin states that the prediction was made before Müller's discovery and was then confirmed.
    4. 4Experiment/Touch and shake responses were repeated on Oxalis, but without formal controls.
    5. 5Replication//Müller, Morren, Hofmeister and Darwin independently report similar movements.
    6. 6Falsifiability//The movements could be incidental effects of general motility, and Darwin says so, but the text does not test whether selection acted on them.

    What would move it up Field or garden experiments measuring whether plants with stronger revolving or touch-sensitivity fare better where support is available, and phylogenetic tests of how climbing arose in each family.

  4. C4InferredC

    Many morphological characters that systematists treat as important (leaf arrangement, floral parts, ovule position) are of little use to the plant. They arose from the laws of growth, spontaneous variation and the nature of the organism, and were not shaped by selection.

    “morphological differences, which we consider as important—such as the arrangement of the leaves, the divisions of the flower or of the ovarium, the position of the ovules, &c., first appeared in many cases as fluctuating variations”

    Evidence offered Documented within-plant and within-species variation in sepal number, placentation, ovary number and seed form under apparently identical conditions, and a comparison with the constancy of hair, feathers and scales in descendants.

    1. 1Observation//The within-plant variation is carefully catalogued, but the lack of fitness effect is not measured.
    2. 2Hypothesis/'Laws of growth' is a label for unknown causes, which limits how precisely the claim can be wrong.
    3. 3Prediction—No prediction is made that could separate neutral from selected variation.
    4. 4Experiment—No test of the fitness effects of these variants is offered.
    5. 5Replication/Many botanists report similar variation, but none tests selection.
    6. 6Falsifiability/Darwin cautions that uses may be undiscovered, which weakens the inference from apparent uselessness.

    What would move it up Fitness measurements (survival, seed set) for the alternative variants, such as petal number or ovule position, in natural populations, together with developmental explanations of the supposed laws of growth.

  5. C5ExtrapolatedD

    Some structures, such as the flatfish's migrated eye and the prehensile tail of American monkeys, owe their origin chiefly to habit and use, with the effects inherited and then strengthened by selection.

    “the transportal of the lower eye of a flat-fish to the upper side of the head, and the formation of a prehensile tail, may be attributed almost wholly to continued use, together with inheritance.”

    Evidence offered Malm's observations that young flatfish twist the lower eye upward and strain the skull, the plasticity of cartilaginous skulls, the lop-eared rabbit, Henslow's harvest mice curling their tails, and the inheritance of the tendency 'judging from analogy'.

    1. 1Observation//Individual behaviour and skull plasticity were observed, but not inheritance of the resulting change.
    2. 2Hypothesis/The claim is loose, and Darwin admits it is impossible to say how much is due to use and how much to selection.
    3. 3Prediction—No risky prediction is made.
    4. 4Experiment—No test shows that a use-induced modification is transmitted.
    5. 5Replication/Malm's observations of related fishes are similar, but they do not address inheritance.
    6. 6Falsifiability—Any structural outcome could be credited to use, selection or both, so no result is ruled out.

    What would move it up Breeding experiments showing that an acquired, use-induced modification of the skull or tail is transmitted to offspring. Later genetics has not supported such inheritance, which cuts against this claim as stated.

  6. C6AssumedE

    Although there is no good evidence for an innate tendency toward progress, a general advance in organisation follows necessarily from selection, because organisation is defined as specialisation of parts and selection favours efficient specialised parts.

    “yet this necessarily follows, as I have attempted to show in the fourth chapter, through the continued action of natural selection.”

    Evidence offered A definition of high organisation as differentiation of parts and an appeal to efficiency, with the argument deferred to chapter four. The same chapter concedes that retrogression occurs in some cases.

    1. 1Observation—Darwin concedes that there is no good evidence of progressive development.
    2. 2Hypothesis/A general advance is predicted, but the concession of retrogression leaves it unclear what would count against it.
    3. 3Prediction—No prediction is stated.
    4. 4Experiment—Nothing is tested.
    5. 5Replication—There is nothing to replicate.
    6. 6Falsifiability/Cases of degradation are acknowledged, but they are treated as exceptions and do not test the general claim.

    What would move it up Phylogenetic data showing that lineages of increasing differentiation do in fact outlast, or leave more descendants than, simpler ones, measured independently of the definition of organisation.

  7. C7InferredC

    Species have arisen mostly by very small steps, and sudden major transformations (Mivart's saltations) are improbable: large variants would be lost by chance or intercrossing, closely graded species abound, and embryology shows no trace of sudden change.

    “That many species have been evolved in an extremely gradual manner, there can hardly be a doubt.”

    Evidence offered Experience that abrupt domestic variants are lost unless breeders preserve them, the many closely allied and doubtful species, similar embryos of birds, bats and quadrupeds, and the argument that the geological record is incomplete.

    1. 1Observation//Graded related species and the loss of abrupt variants under domestication are observed, but not speciation itself.
    2. 2Hypothesis//Gradualism implies specific expectations about intermediates, embryos and variants.
    3. 3Prediction/Few risky predictions are made, and the fossil gaps are handled by invoking an imperfect record.
    4. 4Experiment—No experiment on the tempo of speciation is offered.
    5. 5Replication/Many naturalists report graded species, but this is not independent testing of the tempo.
    6. 6Falsifiability//The alternative of saltation is considered with reasons, though the fossil evidence is explained away conditionally.

    What would move it up Dated, densely sampled fossil lineages and population-genetic data on the size of the variants fixed during speciation, compared against saltational expectations.

Reasoning check

  • Clear case
    Claim C6
    “and natural selection tends towards this end, inasmuch as the parts are thus enabled to perform their functions more efficiently.”

    Here selection is given a goal ('this end'), and the conclusion that progress 'necessarily follows' rests on that. Without agent language only a local statement remains: where a specialised variant is more efficient, it may leave more offspring. That gives no general direction, and Darwin concedes retrogression and defers the argument to chapter four.

  • Possible
    Claim C1
    “That the Pleuronectidæ are admirably adapted by their flattened and asymmetrical structure for their habits of life, is manifest from several species, such as soles, flounders, &c., being extremely common.”

    Here adaptation is read from abundance, so success is evidence of advantage and advantage explains success. Elsewhere the text names independent advantages: Niata cattle cannot browse twigs, giraffes reach higher food, and young flatfish are described by Malm. Measuring fitness differences directly, for example survival by trait value, would settle whether 'advantage' is independent of survival.

  • Possible
    Claim C2
    “it seems to me almost certain that an ordinary hoofed quadruped might be converted into a giraffe.”

    Darwin tells a plausible account of the giraffe's origin and rates it 'almost certain', but no part of the history is directly tested. He offers real evidence: measured proportional variation, the giraffe's feeding advantage and the Niata cattle. Survival data by neck length under dearth, or a fossil lineage, would settle it.

  • Possible
    Claim C4
    “As these variations seem of no special use to the plants, they cannot have been influenced by natural selection.”

    Apparent uselessness is taken as proof of no selection, though Darwin earlier warns against deciding what is useful, and he admits ignorance of the cause. He does offer some evidence: the traits vary widely within single plants under similar conditions, which suggests weak selection. Field fitness measurements of the variants would settle it.

What it does well

  • Darwin separates what selection preserves from where variation comes from: he names individual variation, spontaneous variation, correlated growth, use and disuse, and admits that causes are often unknown.
  • He spells out the mechanism behind the selection language (the highest browsers survive dearths, intercross and leave offspring), so most agent wording can be replaced without losing the argument.
  • He addresses the breeder analogy directly. Unconscious selection needs no foresight or goal, and he notes that nature does not isolate pairs, whereas abrupt variants survive under domestication only by human care.
  • He makes and reports a genuine prediction (nascent revolving and touch-sensitivity in non-climbers) and its confirmation by independent observers.
  • He is candid about limits: he labels conjectures as conjectures, disclaims literal claims about baleen ancestors, concedes that he underrated spontaneous variation, and does not claim to explain why some structures never evolved.

Limits of this evaluation This is one chapter of a longer work, and much support (chapters 2 to 4, Variation under Domestication, the orchid and climbing-plant monographs) is deferred and not counted here. The focus checks otherwise found: agency language was shorthand for a spelled-out mechanism except in the progress passage; the breeder analogy was answered on foresight and isolation; and the sources of variation were named, though their causes were unknown. Later work cuts against the inheritance of use effects and the swamping of rare variants under blending inheritance, a gap this chapter does not address. The cumulative appeal to the 'imperfection of the geological record' is only touched on here and is not rated.