Chapter XII. Geographical Distribution
Charles Darwin · 1872. Read the original ↗
The distribution of living things is shaped by barriers, migration and inheritance with modification, rather than by physical conditions alone. Species arose in single areas and spread by identifiable means, including seed dispersal and the climate shifts of the Glacial period.
The observational and experimental parts (distribution patterns, seed dispersal, glacial traces) are strong and often independently confirmed. The main conclusions, descent with modification and single centres of origin, are inferences to the best explanation from patterns rather than observed or tested events. The selection-based account of northern dominance is only a suspicion without independent tests, and the creation rival is partly dismissed on an unstated premise.
Claims, graded
- C1ObservedB
Differences and similarities among the inhabitants of regions cannot be explained wholly by climate and other physical conditions. Regions with similar conditions can hold very different life, and regions with different conditions can hold related life, and barriers track the differences.
“neither the similarity nor the dissimilarity of the inhabitants of various regions can be wholly accounted for by climatal and other physical conditions.”
Evidence offered Comparisons of the Old and New Worlds, and of Australia, South Africa and South America at similar latitudes. Also the pattern of marine faunas on either side of isthmuses, open ocean and mountain ranges.
- 1Observation///The distributions and the climatic comparisons are directly observed facts, reported from many naturalists.
- 2Hypothesis//The claim is clear and could fail, but 'wholly accounted for' is not given a quantitative test.
- 3Prediction//It predicts that barriers, not climate, track faunal differences, as with the Panama fishes and the open Pacific.
- 4Experiment/Only natural comparisons are used, with no systematic control for climate.
- 5Replication//Darwin notes that almost every author who studied the subject reached the same conclusion.
- 6Falsifiability///The physical-conditions rival is directly set against the data, and similar climates with matching life would have counted against the claim.
What would move it up A quantitative survey across many regions that measures faunal similarity against climatic similarity and barrier strength, with statistical controls.
- C2InferredC
The 'organic bond' linking the life of one region across space and time is inheritance. The dissimilarity between regions comes from descent with modification through variation and natural selection, and only secondarily from physical conditions.
“The bond is simply inheritance, that cause which alone, as far as we positively know, produces organisms quite like each other, or, as we see in the case of varieties, nearly alike.”
Evidence offered Regional types replace one another in close relatives (rheas, South American rodents, island inhabitants resembling the nearest mainland), and the same types appear in the fossils of the same areas. Darwin argues that creation and physical conditions do not explain this pattern as well.
- 1Observation///The geographic patterns of relationship are observed; the descent that supposedly produced them is not.
- 2Hypothesis//It is specific about patterns (islands resemble the nearest mainland, regional types persist through time), though the degree of modification is left open.
- 3Prediction//It anticipates island and regional affinities, but most of the patterns were already known and are being explained after the fact.
- 4Experiment—No controlled test of descent with modification producing regional divergence is offered here.
- 5Replication/Many naturalists report similar patterns, but they are not independent tests of the causal claim.
- 6Falsifiability//Physical conditions and independent creation are compared, but the creation rival is dismissed partly on an unstated premise about what a creator would do.
What would move it up Independent lines of evidence, such as lineage relationships worked out from traits unrelated to geography, that predict where related forms should occur, together with an observed case of regional divergence from a known founding population.
- C3InferredC
Each species was produced in a single area and then migrated as far as its powers and conditions allowed. Apparent exceptions are discontinuities produced by past geographic and climatic change.
“the view of each species having been produced in one area alone, and having subsequently migrated from that area as far as its powers of migration and subsistence under past and present conditions permitted, is the most probable.”
Evidence offered Ranges are usually continuous. Terrestrial mammals are not shared across oceans, though some plants are. Seed-dispersal experiments show means of crossing water, and the Glacial period explains many disjunct ranges.
- 1Observation//Continuity of ranges and the absence of shared mammals across oceans are observed, but origin at a single point is not.
- 2Hypothesis//It is stated clearly, though exceptions can always be attributed to unknown past changes.
- 3Prediction//It predicts no shared mammal species across wide oceans, and none were found.
- 4Experiment/The seed experiments test whether dispersal is feasible, not whether single-origin holds.
- 5Replication/Other naturalists share the view, but the specific test cases are each handled differently.
- 6Falsifiability//Darwin admits many unexplained cases, so there are results that count against the view, but they are set aside on grounds of ignorance.
What would move it up Documented plausible dispersal routes for the remaining unexplained cases (the Kerguelen–New Zealand–Fuegia plants, for example), and independent evidence of where a widespread species first arose.
- C4ObservedB
A substantial fraction of plant species can have their seeds carried across hundreds of miles of sea (and by birds, floating timber or ice) and still germinate. Occasional means of transport are therefore adequate to colonise distant islands.
“the seeds of 14/100 kinds of plants of any country might be floated by sea-currents during twenty-eight days, and would retain their power of germination”
Evidence offered Darwin's own immersion and flotation experiments on 87 and 94 kinds of plants, plus tests of seeds in birds' crops and stomachs and in soil on birds' feet. Martens ran a similar experiment with a better method.
- 1Observation///Germination after immersion, flotation times, and viable seeds from birds and soil were measured directly.
- 2Hypothesis//The claim gives specific numbers, though scaling them to whole floras and 900 miles is flagged as rough.
- 3Prediction/Darwin reports results as surprising, so the tests were exploratory rather than risky predictions.
- 4Experiment///Controlled comparisons of fresh against dried material and of different seed types were made, with a note that the sampling was biased.
- 5Replication//Martens independently repeated the experiment with a better design and found similar results.
- 6Falsifiability//The caveats are open: seeds sank, some orders died, and the estimate is called 'scanty facts'. But actual transoceanic arrivals were not observed.
What would move it up Larger samples and direct records of seeds arriving on new or recently formed islands, to show that dispersal at these rates really occurs and leads to colonisation.
- C5InferredC
The Glacial period explains why identical or closely allied species live on widely separated mountain-tops and in arctic lands. Cold-adapted forms spread south as the climate cooled and were stranded on high ground as it warmed.
“seem to me to explain in so satisfactory a manner the present distribution of the Alpine and Arctic productions of Europe and America”
Evidence offered Observed glacial traces (moraines, scored rocks, erratic boulders) across Europe, North America, the Himalaya, New Zealand and the Andes. Alpine floras are related to the arctic forms lying due north of them.
- 1Observation//The glaciation itself is well observed, but the plant migration that explains the pattern is inferred.
- 2Hypothesis///It is specific: alpine plants should resemble the arctic forms due north, and isolated populations should be modified.
- 3Prediction//The geographic pattern of affinities follows from the model, though much of it was already known.
- 4Experiment/Systematic mapping of glacial deposits serves as a natural test, but there are no direct tests of the migration.
- 5Replication//Many independent observers (Agassiz, Hooker, Haast, Hector, Forbes) found glacial evidence in different regions.
- 6Falsifiability//Independent creation and multiple land-bridges are compared, but later decisive tests of lowland plant occupation are missing.
What would move it up Dated plant remains from glacial-age lowlands showing the alpine species living there. Later pollen records and genetic studies have tended to support such range shifts.
- C6InferredC
Many more species have migrated from the northern to the southern hemisphere than in the reverse direction. This is shown by the present-day asymmetry of the occurrence of northern forms in southern floras.
“many more identical or slightly modified species have migrated from the north to the south, than in a reversed direction”
Evidence offered Reports by Hooker and de Candolle on plants in America and Australia. European species are naturalised in La Plata and New Zealand, while few southern forms are naturalised in Europe despite imports of wool and hides.
- 1Observation//The asymmetry in present-day distribution and naturalisation is reported, but direction of past migration is inferred.
- 2Hypothesis//The asymmetry is stated plainly and can be checked against floras.
- 3Prediction/It follows loosely from the glacial model but is mostly an explanation of a known pattern.
- 4Experiment/The wool and hide imports act as a partial natural test of southern seed supply, but they are not controlled.
- 5Replication//Hooker for America and de Candolle for Australia independently report the same asymmetry.
- 6Falsifiability/Darwin mentions exceptions (Borneo, Abyssinia, the Neilgherries), but alternative causes of the asymmetry are barely tested.
What would move it up Systematic counts of northern and southern native species by region, with controls for seed supply and habitat, and independent dating of arrival.
- C7ExtrapolatedD
The northern asymmetry arises because northern forms, living in larger areas in greater numbers, were advanced by natural selection and competition to a higher 'dominating power' than southern forms. Hence they prevailed when the floras mingled.
“having consequently been advanced through natural selection and competition to a higher stage of perfection, or dominating power, than the southern forms”
Evidence offered Greater land area in the north, and the observed spread of European species over native ones in La Plata and New Zealand, compared with the few southern forms naturalised in Europe. Darwin says 'I suspect', and offers no independent measure of competitive ability.
- 1Observation/Naturalisation outcomes are observed, but the history of selection that supposedly made northern forms more powerful is not.
- 2Hypothesis/'Perfection, or dominating power' is loosely defined and is mostly identified with the outcome it explains.
- 3Prediction/It anticipates the asymmetry but does not state in advance what else should follow, for example from larger areas in the south.
- 4Experiment—No competition or transplant test between northern and southern species is offered.
- 5Replication—No independent test of the causal claim is cited.
- 6Falsifiability/Greater seed supply and other rivals are only partly addressed, and the 'dominance' idea can absorb the exceptions.
What would move it up Independent competition experiments between northern and southern species, a defined measure of competitive ability set before the outcome, and a check on regions where the larger land area is in the south.
Reasoning check
- PossibleClaim C7
“having consequently been advanced through natural selection and competition to a higher stage of perfection, or dominating power, than the southern forms”
'Perfection' is equated with 'dominating power', and dominance is identified by winning, which is the outcome being explained. Darwin does offer independent proxies (larger land area, greater numbers, observed naturalisation), so the circle is not closed. Competition experiments or a measure of dominance fixed before the outcome would settle it.
- PossibleClaim C3
“considering our ignorance with respect to former climatical and geographical changes, and to the various occasional means of transport, the belief that a single birthplace is the law seems to me incomparably the safest.”
Darwin offers real evidence for the explained cases: seed experiments, the glacial migrations, and the lack of shared mammals. For the unexplained remainder, ignorance of past changes is used to leave the single-origin view standing. Documented routes for the remaining cases, or a failure to find any after a thorough search, would settle it.
- PossibleClaim C5
“when in other regions we find the same species on distant mountain-summits, we may almost conclude, without other evidence, that a colder climate formerly permitted their migration across the intervening lowlands”
For Europe and North America the glaciation is independently established and the predicted affinity pattern fits. Extending it to other regions 'without other evidence' infers the cause from the distribution alone, where occasional dispersal might also give the same pattern. Independent glacial evidence and dated plant records at each site would settle it.
- PossibleClaim C5
“It may be suspected that before this flora was exterminated during the last Glacial epoch, a few forms had been already widely dispersed to various points of the southern hemisphere by occasional means of transport”
Some southern species look too distinct to have diverged since the last Glacial period, so Darwin adds an earlier, now-vanished Antarctic flora. The only support is the pattern of southern-confined genera across distant lands, and he flags the idea as a suspicion. Fossil or other dated evidence of that flora would settle it, and later Antarctic fossil finds are consistent with it.
- Clear caseClaim C2
“We cannot maintain that such species have been created alike, in correspondence with the nearly similar physical conditions of the areas”
The argument that similar climates holding dissimilar life count against creation assumes a creator would fit species to climate. That premise about creative intentions is not stated or argued for, though the underlying observation is sound. Without it, the evidence still supports descent with modification as a good explanation but does not rule out creation hypotheses that are not tied to climate.
What it does well
- Darwin runs real experiments on seed flotation, germination after bird digestion, and viable seeds in soil on birds' feet, and he reports failures and sampling biases as well as successes.
- Martens's more realistic repeat of the flotation experiment gives independent confirmation, and Darwin adjusts his own estimate downward in response.
- He compares rivals openly: physical conditions, Forbes's land bridges (rejected on the volcanic composition of islands and depth of seas), and creation at multiple centres.
- The central hypotheses make risky, checkable claims, such as no shared terrestrial mammal species across oceans and alpine floras allied to arctic forms due north.
- He admits large unexplained residues and revises his own earlier opinion in light of Croll's work on alternating glaciation.
- He states that migration and isolation 'in themselves effect' nothing and that some species stay unchanged, so the theory does not predict change everywhere.
Limits of this evaluation This is one chapter of a longer work; the argument for selection, variation and the struggle for existence is given in earlier chapters and is not counted against this section. On the agency-language focus: phrases such as natural selection 'taking advantage' of variability, species 'triumphing' and becoming 'victorious', and 'efficient workshops' recur, but each can be restated in terms of individuals that vary and leave more or fewer descendants ('only so far as it profits each individual'), and the distribution arguments still work. I therefore listed no reification finding. The race-horse comparison is used only to show that many individuals can change at once, not anything that depends on a breeder's foresight, so no false analogy was found. Variation is called 'an independent property' of each species, so selection is not credited with producing it, and no filter-as-source finding was made; where the variation comes from is deferred to earlier chapters. The tautology concern is limited to C7. As a historical work, later findings bear on some claims: plate tectonics and continental drift show Darwin was wrong that the continents remained in nearly the same positions, and later fossils, pollen and genetic data have supported many of his dispersal and glacial range shifts.