Estudos experimentais sôbre a origem do milho

1) It may seem rather strange that, in spite of the efforts of a considerable number of scientists, the problem of the origin of indian corn or maize still has remained an open question. There are no fossil remains or archaeological relics except those which are quite identical with types still existing. (Fig. 1). The main difficulty in finding the wild ancestor- which may still exist - results from the fact that it has been somewhat difficult to decide what it should be like and also where to look for it. 2) There is no need to discuss the literature since an excellent review has recently been published by MANGELSDORF and REEVES (1939). It may be sufficient to state that there are basically two hypotheses, that of ST. HILAIRE (1829) who considered Brazilian pod corn as the nearest relative of wild corn still existing, and that of ASCHERSON (1875) who considered Euchlaena from Central America as the wild ancestor of corn. Later hypotheses represent or variants of these two hypotheses or of other concepts, howewer generally with neither disproving their predecessors nor showing why the new hypotheses were better than the older ones. Since nearly all possible combinations of ideas have thus been put forward, it har- dly seems possible to find something theoretically new, while it is essential first to produce new facts. 3) The studies about the origin of maize received a new impulse from MANGELSDORF and REEVES'S experimental work on both Zea-Tripsacum and Zea-Euchlaena hybrids. Independently I started experiments in 1937 with the hope that new results might be obtained when using South American material. Having lost priority in some respects I decided to withold publication untill now, when I can put forward more concise ideas about the origin of maize, based on a new experimental reconstruction of the "wild type". 4) The two main aspects of MANGELSDORF and REEVES hypothesis are discussed. We agree with the authors that ST. HILAIRE's theory is probably correct in so far as the tunicata gene is a wild type relic gene, but cannot accept the reconstruction of wild corn as a homozygous pod corn with a hermaphroditic tassel. As shown experimentally (Fig. 2-3) these tassels have their central spike transformed into a terminal, many rowed ear with a flexible rachis, while possessing at the same time the lateral ear. Thus no explanation is given of the origin of the corn ear, which is the main feature of cultivated corn (BRIEGER, 1943). The second part of the hypothesis referring to the origin of Euchlaena from corn, inverting thus ASCHERSON's theory, cannot be accepted for several reasons, stated in some detail. The data at hand justify only the conclusion that both genera, Euchlaena and Zea, are related, and there is as little proof for considering the former as ancestor of the latter as there is for the new inverse theory. 5) The analysis of indigenous corn, which will be published in detail by BRIEGER and CUTLER, showed several very primitive characters, but no type was found which was in all characters sufficiently primitive. A genetical analysis of Paulista Pod Corn showed that it contains the same gene as other tunicates, in the IV chromosome, the segregation being complicated by a new gametophyte factor Ga3. The full results of this analysis shall be published elsewhere. (BRIEGER). Selection experiments with Paulista Pod Corn showed that no approximation to a wild ancestor may be obtained when limiting the studies to pure corn. Thus it seemed necessary to substitute "domesticated" by "wild type" modifiers, and the only means for achieving this substitution are hybridizations with Euchlaena. These hybrids have now been analysed init fourth generation, including backcrosses, and, again, the full data will be published elsewhere, by BRIEGER and ADDISON. In one present publication three forms obtained will be described only, which represent an approximation to wild type corn. 6) Before entering howewer into detail, some arguments against ST. HILAIRE's theory must be mentioned. The premendelian argument, referring to the instability of this character, is explained by the fact that all fertile pod corn plants are heterozygous for the dominant Tu factor. But the sterility of the homozygous TuTu, which phenotypically cannot be identified, is still unexplained. The most important argument against the acceptance of the Tunicata faetor as wild type relic gene was removed recently by CUTLER (not yet published) who showed that this type has been preserved for centuries by the Bolivian indians as a mystical "medicine". 7) The main botanical requirements for transforming the corn ear into a wild type structure are stated, and alternative solutions given. One series of these characters are found in Tripsacum and Euchlaena : 2 rows on opposite sides of the rachis, protection of the grains by scales, fragility of the rachis. There remains the other alternative : 4 rows, possibly forming double rows of female and male spikelets, protection of kernels by their glumes, separation of grains at their base from the cob which is thin and flexible. 8) Three successive stages in the reconstruction of wild corn, obtained experimentally, are discussed and illustrated, all characterized by the presence of the Tu gene. a) The structure of the Fl hybrids has already been described in 1943. The main features of the Tunicata hybrids (Fig. -8), when compared with non-tunicate hybrids (Fig. 5-6), consist in the absence of scaly protections, the fragility of the rachis and finally the differentiation of the double rows into one male and one female spikelet. As has been pointed out, these characters represent new phenotypic effects of the tunicate factor which do not appear in the presence of pure maize modifiers. b) The next step was observed among the first backcross to teosinte (Fig. 9). As shown in the photography, Fig. 9D, the features are essencially those of the Fl plants, except that the rachis is more teosinte like, with longer internodes, irregular four-row-arrangement and a complete fragility on the nodes. c) In the next generation a completely new type appeared (Fig. 10) which resembles neither corn nor teosinte, mainly in consequence of one character: the rachis is thin and flexible and not fragile, while the grains have an abscission layer at the base, The medium sized, pointed, brownish and hard granis are protected by their well developed corneous glumes. This last form may not yet be the nearest approach to a wild grass, and I shall try in further experiments to introduce other changes such as an increase of fertile flowers per spikelet, the reduction of difference between terminal and lateral inflorescences, etc.. But the nature of the atavistic reversion is alveadwy such that it alters considerably our expectation when looking for a still existing wild ancestor of corn. 9) The next step in our deductions must now consist in an reversion of our question. We must now explain how we may obtain domesticated corn, starting from a hypothetical wild plant, similar to type c. Of the several changes which must have been necessary to attract the attention of the Indians, the following two seem to me the most important: the disappearance of all abscission layers and the reduction of the glumes. This may have been brought about by an accumulation of mutations. But it seems much more probable to assume that some crossing with a tripsacoid grass or even with Tripsacum australe may have been responsible. In such a cross, the two types of abscission layer would be counterbalanced as shown by the Flhybrids of corn, Tripsacum and Euchlaena. Furthermore in later generations a.tu-allele of Tripsacum may become homozygous and substitute the wild tunicate factor of corn. The hypothesis of a hybrid origin of cultivated corn is not completely new, but has been discussed already by HARSHBERGER and COLLINS. Our hypothesis differs from that of MANGELSDORF and REEVES who assume that crosses with Tripsacum are responsible only for some features of Central and North American corn. 10) The following arguments give indirects evidence in support of our hypothesis: a) Several characters have been observed in indigenous corn from the central region of South America, which may be interpreted as "tripsacoid". b) Equally "zeoid" characters seem to be present in Tripsacum australe of central South-America. c) A system of unbalanced factors, combined by the in-tergeneric cross, may be responsible for the sterility of the wild type tunicata factor when homozygous, a result of the action of modifiers, brought in from Tripsacum together with the tuallele. d) The hybrid theory may explain satisfactorily the presence of so many lethals and semilethals, responsible for the phenomenon of inbreeding in cultivated corn. It must be emphasized that corn does not possess any efficient mechanism to prevent crossing and which could explain the accumulation of these mutants during the evolutionary process. Teosinte which'has about the same mechanism of sexual reproduction has not accumulated such genes, nor self-sterile plants in spite of their pronounced preference for crossing. 11) The second most important step in domestication must have consisted in transforming a four rowed ear into an ear with many rows. The fusion theory, recently revived byLANGHAM is rejected. What happened evidently, just as in succulent pXants (Cactus) or in cones os Gymnosperms, is that there has been a change in phyllotaxy and a symmetry of longitudinal rows superimposed on the original spiral arrangement. 12) The geographical distribution of indigenous corn in South America has been discussed. So far, we may distinguish three zones. The most primitive corn appears in the central lowlands of what I call the Central Triangle of South America: east of the Andies, south of the Amazone-Basin, Northwest of a line formed by the rivers São Prancisco-Paraná and including the Paraguay-Basin. The uniformity of the types found in this extremely large zone is astonishing (BRIEGER and CUTLER). To the west, there is the well known Andian region, characterized by a large number of extremely diverse types from small pop corn to large Cuszco, from soft starch to modified sweet corn, from large cylindrical ears to small round ears, etc.. The third region extends along the atlantic coast in the east, from the Caribean Sea to the Argentine, and is characterized by Cateto, an orange hard flint corn. The Andean types must have been obtained very early, and undoubtedly are the result of the intense Inca agriculture. The Cateto type may be obtained easily by crosses, for instance, of "São Paulo Pointed Pop" to some orange soft corn of the central region. The relation of these three South American zones to Central and North America are not discussed, and it seems essential first to study the intermediate region of Ecuador, Colombia and Venezuela. The geograprical distribution of chromosome knobs is rapidly discussed; but it seems that no conclusions can be drawn before a large number of Tripsacum species has been analysed.

Competição entre megaspórios em milho

The experiments reported were started as early as 1933, when indications were found in class material that the factor for small pollen, spl, causes not only differences in the size of pollen grains and in the growth of pollen tubes, but also a competition between megaspores, as first observed by RENNER (1921) in Oenothera. Dr. P. C. MANGELSDORF, who had kindly furnished the original seeds, was informed and the final publication delayed untill his publication in 1940. A further delay was caused by other circunstances. The main reason for the differences of the results obtained by SINGLETON and MANGELSDORF (1940) and those reported here, seems to be the way the material was analysed. I applied methods of a detailed statistical analysis, while MANGELSDORF and SINGLETON analysed pooled data. 1) The data obtained on pollen tube competition indicate .that there is about 3-4% of crossing-over between the su and sp factors in chromosome IV. The elimination is not always complete, but from 0 to 10% of the sp pollen tubes may function, instead of the 50% expected without elimination. These results are, as a whole, in accordance with SINGLETON and MANGELSDORF's data. 2) Female elimination is weaker and transmission determined as between 16 to 49,5%, instead of 50% without competition, the values being calculated by a special formula. 3) The variability of female elimination is partially genotypical, partially phenotypical. The former was shown by the difference in the behavior of the two progenies tested, while the latter was very evident when comparing the upper and lower halves of ears. For some unknown physiological reason, the elimination is generally stronger in the upper than in the lower half of the ear. 4) The female elimination of the sp gene may be caused theoretically, by either of two processes: a simple lethal effect in the female gametophyte or a competition between megaspores. The former would lead not only to the abortion of the individual megaspores, but of the whole uniovulate ovary. In the case of the latter, the abortive megaspore carrying the gene sp will be substituted in each ovule by one of the Sp megaspores and no abortion of ovaries may be observed. My observations are completely in favor of the second explication: a) The ears were as a whole very well filled except for a few incomplete ears which always appear in artificial pollinations. b) Row arrangement was always very regular. c) The number of kernels on ears with elimination is not smaller than in normal ears, but is incidentally higher : with elimnation, in back-crosses 354 kernels and in selfed ears 390 kernels, without elimination 310 kernels per ear. d) There is no correlation between the intensity of elimination and the number of grains in individual ears; the coefficient; of linear correlation, equal to 0,24, is small and insignificant. e) Our results are in complete disagreement whit those reported by SINGLETON and MANGELSDORF (1940). Since these authors present only pooled date, a complete and detailed analysis which may explain the cause of these divergences is impossible.

Drosophilidae (Diptera) associated to fungi: differential use of resources in anthropic and Atlantic Rain Forest areas

This study investigates the Drosophilidae species associated to fruiting bodies of fungi in forested and anthropized environments of the Atlantic Rain Forest Biome, in south and southeastern Brazil. We collected samples of imagoes flying over and emerging from fruiting bodies of species of five fungi families, in six collection sites. We obtained 18 samples, from which emerged 910 drosophilids of 31 species from the genera Drosophila Fallen, 1823, Hirtodrosophila Duda, 1923, Leucophenga Mik, 1886, Mycodrosophila Oldenberg, 1914, Scaptomyza Hardy, 1849, Zaprionus Coquillett, 1901 and Zygothrica Wiedemann, 1830. The Drosophila species collected on fungi, as well as Zaprionus indianus Gupta, 1970, had previously been recorded colonizing fruits, demonstrating their versatility in resource use. Most of these species belong to the immigrans-tripunctata radiation of Drosophila. Our records expands the mycophagous habit (feeding or breeding on fungi) to almost all species groups of this radiation in the Neotropical region, even those supposed to be exclusively frugivorous. Assemblages associated to fungi of forested areas were more heterogeneous in terms of species composition, while those associated to fungi of anthropized areas were more homogeneous. The drosophilids from anthropized areas were also more versatile in resource use.

Cytogenetic study in natural hybrids of Callithrix (Callitrichidae: Primates) in the Atlantic forest of the state of Rio de Janeiro, Brazil

In the Atlantic forest of Rio de Janeiro, Callithrix aurita (É. Geoffroy in Humboldt, 1812) is a native species vulnerable to extinction and C. jacchus (Linnaeus, 1758) and C. penicillata (É. Geoffroy, 1812) are invasive species. The major threats to the native species are habitat degradation and hybridization, although there are currently no genetic data about natural hybrids available. Previous studies have revealed that species of the Callithrix genus are extremely homogeneous in their karyotypes with the exceptions of the morphology and size of the Y chromosome and its nucleolar organizer region (NOR) banding pattern. Three male marmosets captured in the wild in Guapimirim municipality, Rio de Janeiro, Brazil, considered as possible hybrids between C. aurita and C. jacchus or C. penicillata on the basis of pelage pattern, were cytogenetically studied. Metaphase chromosomes were obtained by using short-term lymphocyte cultures and Ag-NOR staining was performed. The hybrids karyotypes were 2n=46, 14 uni- and 30 bi-armed autosomes, a median size submetacentric X and NOR bearing autosomes, being compatible with that observed for the genus. In the three individuals studied, Y chromosomes were similar to those found for C. aurita, without NORs. The data obtained suggest the involvement of C. aurita in natural hybridization with one of the invasive species. We discuss the possible consequences of this hybridization.

Sôbre a hiperplasia celular no mixoma infectuoso do coelho

Definite hyperplasia of cells occurs in the skin lesions of the infectious myxoma of rabbits, more visible in such stages in which the intercellular basophilic substance is rather scanty (fig. 2). The increase in number of cells is the result of simplified forms of mitosis (modified type of mitosis, pseudoamitosis) which might readily be mistaken for amitosis in their final stages. Budding (figs. 20, 28, 29, 30) as well as constriction of the nucleus (figs. 18, 31, 32), and the formation of giant-cells (figs. 33, 34) are not rare. During the entire process the nuclear membrane does not desintegrate as in typical mitosis. Division of the cytoplasm following division of the nucleus has been demonstrated (fig. 17). Typical mitosis is practically absent. The cells which undergo hyperplasia present remarkable changes in their dimension, shape, and structure. The nucleus and cell-body are considerably enlarged (figs. 6, 7, 8). The shape of the nucleus is modified (figs. 8, 10, 15). Hypertrophy of nuclein, either as an intranuclear network (spireme?, figs. 9, 23), or in the form conspicuous, deeply staining masses which appear not to be homogeneous but to be composed of small particles closely clumped ("mulberries"?, figs. 12, 13, 14, 25, 26) occurs in most cells. While some of these pictures are probably related to necrosis of the cells as started by most of the previous workers, it is lekely that some of them may represent developmental stages of the modified mitosis (pseudoamitosis) here reported. In fact, fine cytological details not ordinarily preserved in necrotic cells (figs. 35, 36, 37) may be demonstrated in the socalled myxoma-cells subtted to approved cytological methods of study (fixation in B-15 and P. F. A.-3, staining in iron-hematoxylin).

Contractions in the 2-wasserstein lenght space and thermalization of granular media

An algebraic decay rate is derived which bounds the time required for velocities to equilibrate in a spatially homogeneous flow-through model representing the continuum limit of a gas of particles interacting through slightly inelastic collisions. This rate is obtained by reformulating the dynamical problem as the gradient flow of a convex energy on an infinite-dimensional manifold. An abstract theory is developed for gradient flows in length spaces, which shows how degenerate convexity (or even non-convexity) | if uniformly controlled | will quantify contractivity (limit expansivity) of the flow.

Semidiscretization and long-time asymptotics of nonlinear diffusion equations

We review several results concerning the long time asymptotics of nonlinear diffusion models based on entropy and mass transport methods. Semidiscretization of these nonlinear diffusion models are proposed and their numerical properties analysed. We demonstrate the long time asymptotic results by numerical simulation and we discuss several open problems based on these numerical results. We show that for general nonlinear diffusion equations the long-time asymptotics can be characterized in terms of fixed points of certain maps which are contractions for the euclidean Wasserstein distance. In fact, we propose a new scaling for which we can prove that this family of fixed points converges to the Barenblatt solution for perturbations of homogeneous nonlinearities for values close to zero.

Análise citológica e cariométrica da ação da colchina sôbre a espermatogênese dos hemípteros

The action of colchicine upon the spermatogenesis of Triatoma infestans, (Hemipt. Heteroptera), has been studied and the different categories of giant spermatids that appear during the treatment have been compared with the nuclear volumes of the whole series of normal spermatogenetic stages. The following facts have been ascertained: 1) 4 hours after the treatment the gonial mitotic metaphases, and the 1st. and 2nd. metaphases of meiosis are stopped. The prophasic stages of meiosis and diakynesis appear to be normal. After 9 days of treatment, all the tetrads are broken in the meiotic metaphases and the cells appear with 44 and 22 chromosomes respectively, scattered in the cytoplasm. 2) At 9 days, practically all spermatogenetic stages have disappeared except for a few cysts of spermatogonia, and practically the whole testicle is full of cysts of spermatozoa and spermatid, with some large zones of necrosis with pycnotic nuclei. The spermatids appear to be of different sizes and the statistical analysis of the nuclear volumes gives a polymodal hystogram with 4 modes, whose volumes are in the ratio of 1:2:4:8. Ripe spermatozoa seem to have a certain volume variability, that has not been possible to analyse quantitatively. All these facts confirm what DOOLEY found in the colchicinized Orthoptera testicle. 3) The caryometric analysis conducted statistically on the normal stages of the spermatogenesis (resting spermatogonia, gonial prophases, leptotene, "confused stage", diakynesis, and spermatid) revealed the following facts: a) Considering the volume of the resting, spermatogonia as 1, their mitotic prophases have a volume of 2. Some rare prophases appear to have a volume of 4 and probably belong to tetraployd spermatogonia normally present in the testicle of Hemiptera. b) The first spermatocyte at the beginning of the auxocitary growth (leptotene) has a volume of 2, which is equal to that of them gonial prophase. It grows further during the "confused stage" and reduplicates, reaching thus the volume of 4. Diakynesis has a rather variable nuclear volume and it is higher than volume 4. This is probably of physico-chemical nature and not a growth increase. c) The spermatid at the beginning of the spermiogenetic process has a volume of 1 which is very constant and homogeneous. 4) These results can be summarized concluding that the meiotic process begins from a spermatogonium at the end of his mitotic interphasic growth (vol. 2) and instead of entering into the mitotic prophase transforms itself into the leptotene spermatocyte. During the diplotene ("confused stage") the volume of the nucleus doubles once more and reaches volume 4. In consequence of the two successive meiotic divisions the spermatid, although having an haploid number of chromosomes, has a nuclear volume of 1, just like the diploid spermatogonium. The interpretation of this strange result probably comes from the existence of the "tertiary split" in the chromosomes of the haploid set, that has been illustrated in the Hemiptera by HUGUES SCHRADER and in Orthoptera by MICKEY and co-workers. The tertiary split indicates that the chromosomes of the haploid set are constituted from almost two chromonemata, and this double constitution corresponds to the double cycle of reduplication that takes place during the spermatogenesis starting from the resting gonia. In Triatoma infestans the tertiary split appears in the chromosomes in the 1st. and 2nd. metaphases and in the diakynesis. In the blocked metaphases at the 9th. day of colchicinization some of the 44 elements scattered in the cytoplasm, show, when properly oriented, the split very clearly. Some new and strange facts revealed by SCHRADER and LEUCHTEMBERGER in Arvelius suggest the possibility of other interpretations of the rhythmic growth in special cases. There appears the necessity of more knowledge about the multiple or simple constitution of the chromosomes in somatic and spermatogonial mitosis.

Preparo de filmes-suporte para microscópio eletrônico

Working with low voltage microscope (R.C.A., EMC-2, of 30KV.) the authors verified that parlodion and Formvar films are quickly destroyed by intense heating under the electron beam. They have tried to employ oxide films, as Al2O3 and SiO, more resistant to heat. Al2O3 films are prepared by anodic oxidation of thin aluminium sheets, under 8 to 10 volts in a 3% ammonium citrate solution and subsequent aluminium dissolution in a O.25% HgCl2 solution. These films are very suitable when prepared with highly pure aluminium of extremely homogeneous surface. Best results were obtained with SiO films, evaporated in high vacuum over Parlodion films mounted on metallic grids. Employing 1 or 1.5 mg of SiOm highly homogeneous and resistant films are obtained, having little inferior transparence than the Parlodion ones. Pure SiO films (1.5 mg) are obtained by elimination of the Parlodion under slow heating until 250°C; they are greatly transparent but little resistant to water, thus beeing indicated in dry preparations. For particles which deposite in a chain-like form around thin fibers, the authors employ the mounting on Parlodion fibers, obtained by heating Parlodion films on microscope grids about 190°C.

Social capital and government in the production of public goods

As a response to the rapidly growing empirical literature on social capital and the evidence of its correlation with government performance, we build a theoretical framework to study the interactions between social capital and government's action. This paper presents a model of homogeneous agents in an overlapping generations framework incorporating social capital as the values transmitted from parent to child. The government's role is to provide public goods. First, government expenditure is exogenously given. Then, it will be chosen at the preferred level of the representative agent. For both setups the equilibrium outcomes are characterized and the resulting dynamics studied. Briefly we include an analysis of the effect of productivity growth on the evolution of social capital. The results obtained caution caution against both the crowding out effect of the welfare state and the impact of sustained economic growth on social capital.

Regional decomposition of CO2 emissions in the world: a cluster analysis

The purpose of this paper is to study the possible differences among countries as CO2 emitters and to examine the underlying causes of these differences. The starting point of the analysis is the Kaya identity, which allows us to break down per capita emissions in four components: an index of carbon intensity, transformation efficiency, energy intensity and social wealth. Through a cluster analysis we have identified five groups of countries with different behavior according to these four factors. One significant finding is that these groups are stable for the period analyzed. This suggests that a study based on these components can characterize quite accurately the polluting behavior of individual countries, that is to say, the classification found in the analysis could be used in other studies which look to study the behavior of countries in terms of CO2 emissions in homogeneous groups. In this sense, it supposes an advance over the traditional regional or rich-poor countries classifications .

Convergence of singular integrals with general measures

We show that L2-bounded singular integrals in metric spaces with respect to general measures and kernels converge weakly. This implies a kind of average convergence almost everywhere. For measures with zero density we prove the almost everywhere existence of principal values.

Orientation of the snail Biomphalaria straminea (Dunker, 1848) in response to light in a situation of selection

The behavioral response of Biomphalaria straminea to light was evaluted in terms of location of the snail in a Y-shaped aquarium in a situation of selection and of the rate (cm/hour) and direction of locomotion under homogeneous 9vertical) or differential (horizontal) lighting upon only one arm of the aquarium. The light source consisted of daylight fluorescent lamps with a spectrum close to that of natural light, with illumination varying from 28 to 350 lux. Analysis of the data showed that all animals, whether in groups or isolated, were attracted to light, although the time needed to approach the light source was 50% shorter for the former than for the latter. The rate of locomotion of B. straminea was 35% higher than observed in B. glabrata and 51% higher than that observed in B. tenagophila studied under similar conditions. The results are discussed in terms of social factors and geographical distribution of the three species.

Modelo de un grupo de robots colaboradores cazadores

El nombre d'aplicacions dels microrobots en biomedicina creix a mesura que el seu desenvolupament avança. Entre elles hi ha les consistents a examinar cèl·lules amb microrobots cooperants. En aquest treball es presenta un prototip a escala d'aquest problema, convenientment simplificat: dos robots tracten d'agafar una pilota que representa la cèl·lula que s'examina. Com a resultat, s'ha obtingut un algorisme deliberatiu per a la resolució d'aquest problema amb robots homogenis.

Effects of selection and drift on G matrix evolution in a heterogeneous environment: a multivariate Qst-Fst Test with the freshwater snail Galba truncatula.

Unraveling the effect of selection vs. drift on the evolution of quantitative traits is commonly achieved by one of two methods. Either one contrasts population differentiation estimates for genetic markers and quantitative traits (the Q(st)-F(st) contrast) or multivariate methods are used to study the covariance between sets of traits. In particular, many studies have focused on the genetic variance-covariance matrix (the G matrix). However, both drift and selection can cause changes in G. To understand their joint effects, we recently combined the two methods into a single test (accompanying article by Martin et al.), which we apply here to a network of 16 natural populations of the freshwater snail Galba truncatula. Using this new neutrality test, extended to hierarchical population structures, we studied the multivariate equivalent of the Q(st)-F(st) contrast for several life-history traits of G. truncatula. We found strong evidence of selection acting on multivariate phenotypes. Selection was homogeneous among populations within each habitat and heterogeneous between habitats. We found that the G matrices were relatively stable within each habitat, with proportionality between the among-populations (D) and the within-populations (G) covariance matrices. The effect of habitat heterogeneity is to break this proportionality because of selection for habitat-dependent optima. Individual-based simulations mimicking our empirical system confirmed that these patterns are expected under the selective regime inferred. We show that homogenizing selection can mimic some effect of drift on the G matrix (G and D almost proportional), but that incorporating information from molecular markers (multivariate Q(st)-F(st)) allows disentangling the two effects.