Studies on pattern regulation in Hydra: IV. The effect of colcemide and puromycin on polarity and regulation

Regulation in hydra is polarized, the new hypostome always forming at the distal end of an isolated piece. Polarity is rigorously maintained in such pieces but can be altered in graft combination (Peebles, 1900; King, 1901; Morgan, 1901; Browne, 1909; Goetsch, 1929; Tardent, 1960; Webster, 1966b).

Previous investigations (Webster, 1966a, b; Webster & Wolpert, 1966) have suggested that polarized regulation occurs as a result of the interaction of three factors : time for hypostome determination, inhibition of hypostome determination and threshold for inhibition. All these factors appear to be graded along the axis. There is some evidence that time for hypostome determination and threshold for inhibition are closely linked factors (Webster, 1966b) and in many experiments it is not possible to distinguish between them. For this reason they are referred to throughout this paper as ‘time-threshold’ factors or properties.

If polarized regulation does result from interaction between the above factors, then experimental alteration of polarity would be expected to depend upon interference with these factors or with their interaction. Experimental alteration of polarity in graft combinations has previously been discussed in such terms (Webster, 1966b) but no attention has been devoted to experimental alteration by other means. This is the subject of the present paper.

A large body of literature exists on the effect of various environmental factors (chemical agents, temperature, electric current, etc.) on regulation and polarity (see e.g. Child, 1941). In some hydroids, but not hydra, polarity is normally rather labile and can be influenced by a variety of environmental factors. Tubularia is a good example (literature reviewed by Child, 1941; Tardent, 1960, 1963). Corymorpha is a hydroid which often shows bipolar regulation in short pieces and the frequency of such forms can be increased by exposing the pieces to inhibitory agents (cyanide, alcohol, low temperature, etc.) for a short time after isolation. In the majority of such experiments there has been no detailed analysis of the effect of such agents at either a biological or a biochemical level. It has usually been assumed that the alteration of polarity has been produced as a result of interference with some sort of axial gradient which ‘controls’ polarity but little convincing evidence has been advanced to support this view.

It is surprising that little work along these lines has been carried out with hydra. Lesh & Burnett (1964,1966) have recently demonstrated that a substance is present in homogenates of H. pirardi and H. viridis which when applied to isolated pieces of H. pirardi is capable of altering polarity. Lentz (1965), using slightly different techniques, has made a similar observation. All these workers have interpreted their results in terms of Burnett’s (1961) hypothesis of control of distal regeneration by ‘growth stimulating’ and ‘growth inhibiting’ substances.

The basis for the experiments described in this paper is the work of Flickinger (Flickinger, 1959; Flickinger & Coward, 1962), originating from the observations of Kanatani (1958) that brief treatment of planaria with colcemide (desacetylmethylcolchicine) causes bipolar regulation in isolated pieces. Flickinger showed that there was a gradient in the rate of incorporation of labelled amino acids into protein along the body of the animal, and demonstrated that treatment with colcemide or chloramphenicol flattened or abolished this gradient. Thus the morphological effect of colcemide could be related to a specific biochemical effect.

In this paper experiments are described on the effect of colcemide on polarized regulation in various isolated regions of hydra. The effect of this substance on hypostome formation and the properties of the determined and ‘adult’ hypostome are also considered. Experiments have also been carried out to determine the effect of colcemide on those properties (‘time-threshold’ properties) which are believed to control polarity. Similar experiments have been carried out using puromycin, also an inhibitor of protein synthesis (Yarmolinsky & de la Haba, 1959), since preliminary experiments indicated that chloramphenicol —effective on planaria—is very toxic to hydra.

It must be emphasized that the primary purpose of the experiments was to investigate how these chemical agents affect polarity at a biological level, and to consider the results in terms of concepts of control which have been developed to explain normal regulation. No attempt is made to discuss the possible biochemical effects of these substances and no assumptions are made that they are acting in any particular way at a biochemical level. Such consideration must await the completion of biochemical investigations which are still in progress.

Hydra littoralis were used for all experiments. Details with regard to selection of animals, treatment during experiments, etc., have been given in Webster & Wolpert (1966).

Colcemide (a gift from Ciba Laboratories, Horsham, Sussex) and puromycin dihydrochloride (Nutritional Biochemicals Corp.) were dissolved in ^‘M’ solution and animals were treated in large volumes of the solution at 26 °C. After treatment they were washed three times in ‘M’.

Different regions of the axis were isolated as follows: digestive zones by cuts just proximal to the ring of tentacles and just distal to the budding zone ; peduncles by a cut just proximal to the budding zone. Hypostomes and tentacles or peduncles were removed as in the isolation of the digestive zone. Animals were allowed to extend maximally before cutting.

Transplantation experiments to investigate hypostomal or subhypostomal properties were performed as previously described (Webster & Wolpert, 1966; Webster, 1966a,b).

Colcemide (5 × 10⁻⁵ M, 20 μg/ml) inhibits reconstitution of all missing parts. Hydra placed in colcemide after removal of hypostome and tentacles or peduncle and basal disc show no signs of reconstituting these missing regions within 48 h. Untreated control animals reconstitute both distal and proximal regions within 36 h at 26 °C. Colcemide-treated animals, after removal from colcemide, reconstitute missing regions within 48-120 h as described below.

Hydra treated with colcemide react in a characteristic manner. After 24 h all animals are extremely swollen and if punctured with a needle deflate like a pricked balloon, often spewing out masses of loose endoderm cells. Many animals are spherical at 24 h, but after 48 h most have assumed an irregular shape.

The polarity of hydra which have had hypostome and tentacles removed is usually unrecognizable after 24 h in colcemide, but the body wall of spherical animals often shows a characteristic pattern, being colourless and transparent at the poles as compared with the equator which is dense, pink and translucent. In some cases this pattern can be seen to represent the original linear axis, as indicated by debris characteristic of the basal disc still attached to one pole.

If the hypostome and tentacles are not removed, but the animals are wounded by a small lateral incision or by the removal of proximal regions, the original polarity is still recognizable after 24 h, but the tentacles are considerably reduced in size ; after 48 h the tentacles have disappeared completely and polarity is in general no longer recognizable (Plate 1, fig. 1). The body of such animals often has a characteristic shape; longitudinal depressions appear so that when viewed from one pole the animals have a lobed appearance, the number of lobes being generally equal to the original number of tentacles (five or six). Buds at a medium stage of development often complete development while still in colcemide and may separate from the parent.

All colcemide-treated animals rapidly become enclosed in a gelatinous envelope which becomes thicker during the period in colcemide and after removal. Microscopic examination reveals this envelope to consist mainly of cytolysed cells, particularly cnidoblasts with nematocysts in all stages of differentiation. It is interesting to note that when hydra subsequently reconstitute tentacles these are entirely or partially deficient in nematocysts in about 50 % of the animals.

Puromycin is toxic and concentrations of 1·1-1·7 x 10⁻¹ M (50-75 μg/ml) were the highest that could be employed. Even at these concentrations a proportion of the animals was killed after 24-48 h treatment. Reconstitution was not completely inhibited, and after 24 h some animals showed signs of reconstituting missing peduncles and tentacles.

Puromycin does not produce the striking morphological changes seen in colcemide-treated animals. Intact hydra placed in 1·1 × 10⁻⁴ M puromycin show signs of tentacle and peduncle regression after 24 h, as do those placed in colcemide, but even after 48 h these regions have not completely disappeared and most animals are still axiate with recognizable polarity. Buds often continue to develop and new ones may arise while the animals are still in puromycin.

After removal from puromycin, tentacle reconstitution occurs fairly quickly— within 24-48 h—but missing proximal regions are often reconstituted very slowly, particularly in animals which have been treated for 48 h.

Animals which are wounded or have had proximal regions removed (i.e. basal disc and peduncle) and are treated with colcemide for up to 48 h and then transferred to ‘M’ reconstitute the missing regions and are exclusively monopolar in form. They possess one distinct group of tentacles, located at one end of a more or less linear axis (Table 1).

When animals which have had the hypostome and tentacles removed (or these regions together with peduncle and basal disc) are treated with colcemide for 24-48 h and are then transferred to ‘M ‘, some show signs of tentacle reconstitution within 48 h. Most of these reconstituting animals possess one localized group of 4-6 tentacles in one region of a more or less linear axis ; these become typical monopolar forms. The remaining animals, which in general reconstitute tentacles more slowly, include a significant number which show not one localized group of tentacles but two or more distinct groups, or alternatively, in a few cases, an apparently random arrangement of isolated tentacles on an irregularly shaped body. Many of these animals regulate within 48 h. For example, two adjacent but apparently distinct groups of tentacles may become confluent and a typical monopolar animal is formed, often with branched tentacles if the latter have fused at their bases. Alternatively two widely separated groups of tentacles will each acquire a distal axis and a Y-shaped, V-shaped or linear bipolar form will result (Plate 1, figs. 2-4). Some animals, particularly those showing an apparently random tentacle distribution, show little sign of regulation and retain their irregular appearance for as long as 14 days.

It must be emphasized that in general only distal structures, i.e. tentacles and distal axes, are involved initially in multiplication. In only eight animals were multiple basal discs or peduncles observed and careful study indicated that such forms usually arose when the tentacle group formed in the middle of a long piece, thus leaving two free ends which became peduncles and basal discs (Plate 1, fig. 6).

The further fate of animals possessing multiple distal structures is variable. As noted above, extremely irregular animals usually undergo little change; Y-shaped bipolars usually develop a basal disc at the foot of the Y within 48 h of reconstituting tentacles (Plate 1, fig. 2) and subsequently behave like bipolars produced by grafting a hypostome into the digestive zone of a host animal, i.e. the two axes show no sign of separating for up to 14 days. V-shaped or linear bipolars subsequently form a basal disc at the apex of the V or in the middle of the linear axis (Plate 1, fig. 4), and in many cases separation of the two axes occurs within 7-14 days. The behaviour of these forms is analogous with that of bipolars produced by grafting two hydra together with opposite polarity.

The number of animals showing multiplication of distal structures is affected by both the size of the treated piece and the duration of colcemide treatment (Table 2). Treatment of hydra minus hypostome and tentacles for 24 h usually results in a small number (12 %) of the animals reconstituting multiple distal structures which nearly all regulate to form Y-shaped bipolars with two distinct groups of tentacles, two distal axes and a common proximal axis and basal disc. Increasing the time of treatment of animals minus hypostome and tentacles to 48 h results in an increase in the number of forms showing multiplication of distal structures (22 %), but again the majority regulate to form Y-shaped bipolars. Reducing the size of the piece by employing isolated digestive zones and treating for 24 h results in an increased number of animals with multiple distal structures (28 %), and these in general regulate to form V-shaped or linear rather than Y-shaped bipolars. Treatment of isolated digestive zones for 48 h with colcemide results in greatly retarded tentacle reconstitution; some animals take up to 5-6 days before producing tentacles, but most animals eventually do so. More than 40 % of these animals show multiple tentacle sites and in a few cases they are apparently randomly distributed over an irregularly shaped body. Regulation to produce distinct distal axes does not usually take place in these very irregular forms. Animals which show less drastic alteration regulate to form V-shaped or linear bipolars.

In contrast to the behaviour of isolated digestive zones, peduncles when isolated and treated with colcemide for 24 h reconstituted to produce exclusively monopolar forms.

When intact animals are treated with colcemide for 24 h, removed from colcemide and the digestive zone isolated (i.e. hypostome and tentacles, budding zone, peduncle and basal disc removed) they reconstitute to produce exclusively monopolar forms (Table 3 (1)). Animals which are treated with colcemide for 24 h after removal of the peduncle and basal disc and then have the hypostome and tentacles removed subsequent to colcemide treatment behave in a similar fashion; out of fifty animals treated, only one reconstituted to form a linear bipolar with a tentacle group at each end of the long axis (Table 3 (2)).

It should be emphasized that there are certain technical difficulties in this experiment. As described above, animals which have been treated with colcemide for 24 h are considerably swollen and have partially or totally resorbed their tentacles. For this reason it is extremely difficult to be certain that all the hypostome and tentacles have been removed, since these regions are distorted and spread over a considerable area of one pole of the spherical animal. In an effort to ensure removal of all of these regions rather large amounts of tissue were usually cut away and many of the pieces were therefore considerably smaller in size than in the other experiments. This may account for the fact that a large number of animals failed to reconstitute tentacles in one of these experiments which employed a smaller piece (animal minus peduncle and basal disc) to start with. The single animal showing multiple distal structures may therefore simply be a result of a failure to remove completely all of the hypostome and tentacles, in which case the animal would be expected to reconstitute in a similar manner to those from which the hypostome and tentacles are not removed subsequent to colcemide treatment.

Isolated digestive zones when treated for 24-48 h with 1·1 × 10⁻⁴ M puromycin reconstitute to form typical monopolar hydra (Table 4). One animal from the 24 h group produced a secondary distal axis bearing a single tentacle (very similar to a type 3 induction—Webster & Wolpert, 1966) giving rise to a Y-shaped bipolar.

Treatment of isolated proximal and distal halves of the digestive zone with a slightly higher concentration of puromycin (1·7 × 10⁻⁴ M) resulted in a small increase in the frequency of bipolar reconstitution (Table 4).

Bipolars produced from the distal half of the digestive zone were all Y-shaped of exactly the same type as the one described above. The single one produced from the proximal half was a linear bipolar with two distinct tentacle groups at opposite ends of the long axis.

Thus, although puromycin has similar effects to colcemide in causing treated animals to reconstitute multiple distal structures, it is definitely not as effective as the latter substances. Both a higher concentration of chemical and a smaller size of treated piece are necessary before animals with multiple distal structures are produced in significant numbers.

The effect of colcemide on the formation of a new hypostome was tested by transplanting the distal tip of a reconstituting animal to the digestive zone of an intact host hydra (Webster & Wolpert, 1966). Animals which had been reconstituting in colcemide (5 × 10⁻⁵ M) for 8-9 h showed no signs of hypostome formation from the subhypostomal region, though the majority of control animals possessed determined hypostomes at this time (Table 5).

The organizing ability of the hypostome was unaffected by colcemide treat ment. Fragments of hypostome taken from hydra which had been treated (in the absence of peduncle and basal disc) for 16-17 h induced secondary axes in the majority of the hosts to which they were transplanted (Table 5). The morphological changes and the consequent obliteration of visible polarity which result from colcemide treatment prevent this experiment being performed on animals which have been treated for longer periods.

Subhypostomal regions were isolated and allowed to reconstitute in ‘M’ for 8 h. At this time, transplantation of pieces to the digestive zone of intact host hydra indicated that a determined hypostome was present in the majority of cases (Table 6). The remaining pieces were transferred to colcemide (5 × 10 ⁵ M) or puromycin (1·1 × 10⁻⁴ M) for 16 h, when transplantation experiments were again performed. Treatment for this time in either substance does not result in irreversible changes or death; washed pieces transferred to ‘M’ all reconstituted tentacles within 24-48 h. It is important to note that about 50 % of the pieces which had been treated with puromycin possessed small tentacle buds at the time of transplantation—clear evidence of the presence of a functional hypostome.

The results shown in Table 6 indicate that colcemide treatment has no effect on the determined hypostome, which induced secondary axes in the majority of cases. This confirms the results previously obtained with the ‘adult’ hypostome. Pieces treated with puromycin, however, induce secondary axes in a very few cases only, the majority of the grafts being absorbed. This indicates that puromycin treatment interferes with the ability of the determined hypostome to resist the influence of the factors which bring about absorption following transplantation (Webster, 1966b). It is interesting that one puromycin-treated graft which did not induce was not absorbed, but transformed into a small peduncle complete with basal disc, a striking indication of a change in properties to those of more proximal regions.

Freshly isolated subhypostomal regions were treated with colcemide (5 × 10⁻⁵ M) or puromycin (1·1 × 10⁻⁴ M) for 9h and then transplanted to the digestive zones of host hydra from which the hypostome and tentacles had been removed—the usual test for subhypostomal ‘time-threshold’ properties (Webster, 1966 a, b). Treatment for this time does not produce irreversible changes or injury; washed pieces transferred to ‘M’ all reconstituted tentacles within 24 h.

Puromycin treatment resulted in all the grafts being absorbed. In the case of grafts treated with colcemide, only one case of induction of a secondary axis was observed. Control grafts (untreated) made immediately after isolation behaved in the usual way and produced secondary axes in the majority of cases (Table 7).

The results indicate that treatment of the subhypostomal region with colcemide or puromycin changes the ‘time-threshold’ properties of this region to those characteristic of more proximal regions (Webster, 1966 a, b). The results also, of course, confirm that colcemide inhibits hypostome formation. Whether puromycin acts in a similar fashion cannot be determined since, as the previous experiment showed, puromycin intereferes with the resistance of the hypostome to factors bringing about absorption.

The results show that colcemide treatment of hydra from which the hypostome has been removed can bring about an alteration of polarity such that multiple distal structures, i.e. more than one hypostome, can form in a single animal. The alterations observed were variable and resulted in the production of a range of forms, from animals showing bipolar organization to those in which hypostomes were more or less randomly distributed. Pretreatment with colcemide followed by removal of the hypostome resulted in only one animal reconstituting more than one hypostome. In no case did such forms arise if the original hypostome was not removed. Treatment with puromycin in the absence of the hypostome was much less effective in altering polarity as judged by subsequent reconstitution than was treatment with colcemide. The effects of colcemide and puromycin on ‘physiological’ polarity were paralleled by their morphological effects ; visible polarity was completely obliterated in colcemide treated animals but persisted in those treated with puromycin.

The transplantation experiments indicated that hypostome formation was inhibited by colcemide but that this substance did not irreversibly affect the determined hypostome or the ‘adult’ hypostome, which retained their organizing properties. The results of puromycin treatment are curious. Hypostome formation was not inhibited completely, as indicated by the fact that some animals reconstituted tentacles while in puromycin. However, the hypostome which was formed did not possess the normal resistance to absorption, as shown by the transplantation experiments. Both colcemide and puromycin changed the ‘time-threshold’ properties of the subhypostomal region to those characteristic of more proximal regions.

The simplest way of explaining the effects of colcemide and puromycin on regulation is to postulate that treatment with these substances flattens or abolishes the disto-proximal axial gradient in ‘time-threshold’ properties (Webster, 1966a), so that when released from inhibition distal regions have no advantage over proximal regions as regards hypostome formation. The results of the experiments on the subhypostomal region are direct evidence that treatment with either substance changes the ‘time-threshold’ properties of this region to those characteristic of more proximal regions, i.e. regions which are lower on the gradient. In order that the gradient be flattened, it is clear that distal regions must be affected more than proximal regions. No evidence is available on this point, though the fact that colcemide treatment of isolated peduncles did not result in the formation of multiple distal structures may indicate that proximal regions are less affected. This is also suggested by the fact that animals from which the hypostome and tentacles only have been removed (i.e. animals possessing a peduncle) produced mainly Y-shaped bipolars with a hypostome at each end of the digestive zone. Kanatani (1958) observed that in planaria bipolar reconstitution was more frequent from anterior than from posterior regions.

Consider what would happen when an animal whose hypostome has been removed is placed in colcemide. New hypostome formation is prevented and the level of inhibition will fall, presumably throughout the animal. The gradient in ‘time-threshold’ properties is flattened. On removal from colcemide, all regions will begin forming hypostome, and those with the ‘highest’ ‘time-threshold’ properties will succeed and inhibit those with ‘lower’ properties. The final result will be variable and animals will form one or more hypostomes, depending on how the gradient has been altered and how the developing hypostomes influence each other. A hypostome, once formed, will immediately begin to raise the level of inhibition in its immediate neighbourhood (Webster, 1966b). It is probable that any other hypostome which forms will be as far away from the first as possible, since we know that the effectiveness of inhibition decreases with distance (Webster, 1966a). For example, in an isolated digestive zone the new hypostomes will form at opposite ends and a linear bipolar will result.

When the original hypostome was left in place during colcemide treatment and subsequently removed, multiplication of distal structures was observed in only one animal out of those which reconstituted. It will be remembered that there were technical difficulties in this experiment, and the possibility that the hypostome was not completely removed in all animals makes it unwise to conclude that this procedure is less effective in altering polarity. However, it is possible that leaving the hypostome in situ during treatment tends to counteract the action of colcemide, and this is consistent with the idea that the hypostome plays some role in controlling the gradient in ‘time-threshold’ properties (Webster, 1966b).

In those experiments in which the hypostome was left in place both during and subsequent to colcemide treatment no multipolar forms were produced. Even if the gradient is flattened in these cases, we have seen that the determined hypostome is not irreversibly affected by colcemide treatment, so that, on removal from the substance, a functional hypostome is present which can presumably inhibit hypostome formation elsewhere in the animal. Even if the inhibitory action of the hypostome is impaired by colcemide treatment, suppression of new hypostome formation will occur at a low level of inhibition because of the depression of thresholds.

Puromycin is much less effective than colcemide in causing the production of multipolar forms, even though it has a similar effect to the latter substance on ‘time-threshold’ properties. This is probably due to the fact that puromycin does not inhibit the formation of a hypostome in all animals. An animal (minus hypostome) when placed in puromycin may develop a new hypostome which, although abnormal in resistance to absorption when transplanted, is adequate to inhibit further hypostome formation in situ. Thus reconstitution will be monopolar.

It is interesting to note that colcemide treatment produced about twice as many multipolar forms from isolated digestive zones as compared with larger pieces (animals minus hypostome and tentacles only). Similar results were obtained with puromycin. This seems to be a general feature of bipolar reconstitution. For example, in marine hydroids the frequency of spontaneous bipolar forms is higher in small pieces than in large pieces (Child, 1941). A completely satisfactory explanation of this result is not at present possible. It is clear that the two ends of a small piece will have fairly similar ‘time-threshold’ properties so that it is easier for them to become identical than for the two ends of a large piece. However, this explanation will not account for the difference in numbers between small and large pieces, unless the additional assumption is made that a hypostome has a greater tendency to arise from a cut surface (i.e. an end) than from an intact area.

Although colcemide and puromycin have similar effects on the ‘time-threshold’ properties of the subhypostomal region, their effects on hypostome formation and the determined hypostome are strikingly different. Colcemide appears to block hypostome formation completely, but the determined hypostome is resistant to and is not irreversibly affected by colcemide treatment. The behaviour of the determined hypostome with respect to colcemide is reminiscent of its behaviour with respect to the ‘natural’ inhibitor of hypostome formation (Webster, 1966a).

The effect of puromycin is extremely interesting. This substance does not appear to inhibit hypostome formation since organizing properties may be acquired in some cases (as judged by the production of tentacles). However, it seems to interfere with the resistance to absorption of the determined hypostome, and presumably with the acquisition of resistance during hypostome formation. This suggests very strongly that organizing ability and resistance to absorption are quite distinct properties. If the resistance to absorption normally displayed by the determined hypostome is dependent on resistance to inhibition and therefore upon high threshold, the fact that organizing properties can be acquired and retained in the absence of resistance suggests that the observed rise in threshold during hypostome formation (Webster, 1966 b) might not be a necessary part of the process of hypostome formation. This has important consequences for our understanding of how release from inhibition occurs during normal regulation. If hypostome formation can occur without rise in threshold, then release from inhibition must result from a fall in the level of inhibition.

If the resistance to absorption displayed by the hypostome is dependent on threshold, then its threshold properties must be less labile than those of non-hypostomal regions since we have seen that subhypostomal threshold properties are altered by both colcemide and puromycin but hypostomal resistance is unaffected by colcemide. It is, of course, possible that higher concentrations of colcemide might interfere with hypostomal resistance. It is known that the threshold properties of non-hypostomal regions are unstable and change as a consequence of isolation or transplantation. The hypostome, however, retains its characteristic properties following transplantation and for as long as 7 days after isolation (Webster, 1966 b, and unpublished observations).

An alternative explanation is possible for the different results obtained with colcemide and puromycin if it is assumed that colcemide has no direct effect upon ‘time-threshold’ properties and that these change as a result of the inhibition of hypostome formation. It is known that the ‘time-threshold’ properties of a region are determined by the position of the region on the linear axis in relation to the hypostome. Changing the position of a region so that it is nearer the hypostome ‘raises’ the ‘time-threshold’ properties; moving it farther away from the hypostome ‘lowers’ the ‘time-threshold’ properties. These changes only occur if the level of inhibition remains above the threshold so that hypostome formation is prevented (Webster, 1966b). It seems reasonable to postulate that if a region is removed to an ‘indefinite’ distance from a hypostome, i.e. isolated, and at the same time prevented from forming hypostome by colcemide, then ‘time-threshold’ properties would ‘fall’. This explanation is consistent with the observations that colcemide has little effect on the polarity of animals treated prior to hypostome removal and that the determined hypostome is resistant to treatment. Puromycin, on the other hand, would be supposed to have a direct effect on ‘time-threshold’ properties and hence could affect both hypostomal and non-hypostomal regions. Its relative inefficiency at bringing about alteration of polarity would result from its failure to inhibit hypostome determination. This explanation could be subjected to experimental test by examining the ‘time-threshold’ properties of regions of animals treated with colcemide in the presence of the hypostome. It might be mentioned that in planaria polarity can be altered in animals treated prior to head removal (Kanatani, 1958; Flickinger & Coward, 1962).

The present results are of interest in relation to those of Lesh & Burnett (1964, 1966) and Lentz (1965). These workers demonstrated that a substance was present in homogenates of hydra which when applied to isolated pieces of the digestive zone for 4 h was capable of altering polarity and producing animals with multiple distal structures. Lesh & Burnett have claimed that this substance is identical with the so-called ‘growth-stimulating principle’, postulated by Burnett (1961) to account for distal regeneration in hydra, and also that the substance is ‘responsible for polarity’. The forms produced as a result of treatment with the homogenate appear to be identical with those produced in the present experiments as a result of colcemide treatment. Colcemide is an inhibitor of protein synthesis in planaria (Flickinger, 1959) and of mitosis in a variety of organisms (Schär, Loustalot & Gross, 1954) including hydra (Webster, 1964); it can therefore fairly be described as a ‘growth-inhibiting substance’. The fact that it appears to have identical effects to the substance present in hydra homogenates suggests that the conclusion that the latter is a ‘growth-stimulating substance’ is unjustified and also suggests that ‘growth’ plays no role in the control of polarity. More important, colcemide inhibits hypostome formation and the alterations of polarity which are produced (involving effects on regional ‘time-threshold’ properties) take place in the absence of a hypostome. In other words, hypostome formation as such is not a necessary feature of any alteration in polarity but merely reveals at a later stage that such alterations have taken place. In the experiments of Lesh and Burnett alteration of polarity is said to occur as a result of the actual stimulation of hypostome and tentacle formation. It is worth noting that these workers treated isolated pieces for only 4 h. If the substance they are dealing with is really ‘growth-stimulating’ then continuous treatment during regeneration should have the same or a greater effect than brief exposure. Also, the substance should be effective irrespective of the presence or absence of the hypostome. These critical experiments do not appear to have been done. In the light of these criticisms it must be concluded that the significance for normal regulation of the substance present in hydra homogenates is obscure.

1. L’action de la colcémide et de la puromycine sur la polarité et la régulation chez l’hydre a été étudiée au niveau biologique au moyen de techniques d’isolement et de transplantation.

2. Le traitement à la colcémide d’une hydre dont on a coupé l’hypostome peut déterminer un changement de polarité tel que de multiples structures distales (hypostome et tentacules) peuvent se former sur un seul animal. De telles formations n’apparaissent pas si l’hypostome primitif n’est pas supprimé. Le traitement à la puromycine affecte beaucoup moins la polarité.

3. Des expériences de transplantation montrent que la formation de l’hypo-stome est inhibée par un traitement à la colcémide mais que cette substance n’affecte pas irréversiblement l’hypostome déterminé ou l’hypostome ‘adulte’ qui conservent leurs propriétés organisatrices.

4. Le traitement à la puromycine n’inhibe pas complètement la formation de l’hypostome puisque quelques animaux reconstituent des tentacules pendant ce traitement. Cependant, l’hypostome qui est formé ne possède pas la résistance normale à l’absorption qui suit la transplantation. Ce résultat peut indiquer que les facteurs responsables des propriétés organisatrices de l’hypostome (c’est-à-dire l’induction des tentacules) sont distincts de ceux qui contrôlent la résistance à l’absorption.

5. La colcémide et la puromycine transforment certaines propriétés importantes de la région sous-hypostomale en propriétés caractéristiques des régions plus proximales. On pense que ces propriétés ‘contrôlent’ la polarité.

6. Les effets de la colcémide et de la puromycine sur la polarité et la régulation sont discutés en fonction de leur action sur un système de contrôle qui a été précédemment postulé pour expliquer la régulation polarisée chez l’hydre. Ce système implique des gradients axiaux dans le temps pour la détermination de l’hypostome, l’inhibition de la formation de l’hypostome et le seuil de l’inhibition. Les résultats expérimentaux peuvent être interprétés comme une conséquence de l’action directe ou indirecte de ces substances sur les gradients axiaux.

I should like to thank Dr L. Wolpert for helpful discussions during the course of this work, and the Agricultural Research Council for a Postgraduate Research Studentship.