In Vitro Studies on the Environmental Biology of Goniodes Colchici (Denny) (Mallophaga:Ischnocera): III. The Effects of Temperature and Humidity On the Uptake of Water Vapour

The insect order Mallophaga consists of obligate ectoparasites of birds and mammals. Interest in this group centres upon two aspects of their physiology not commonly met with in other insects. First, there is the ability of insects contained in the suborder Ischnocera to live exclusively on a diet of hair or feathers. Secondly, and most important with regard to the present paper, is the well-established observation that these insects are dependent upon temperatures in the 30–40 °C range for survival and reproduction (Kellog, 1896; Barber, 1921; Wilson, 1934; Matthysse, 1944; Scott, 1952). Goniodes colchici (Denny) is no exception to this pattern of temperature dependence (Williams, 1970 a).

Since these animals survive in temperatures much higher than ambient, and since their water intake must be low, the water balance of the ischnoceran lice must be closely regulated. It has been demonstrated that G. colchici is not able to regulate its rate of water loss more efficiently at temperatures in the 30–40 °C range, but rather that its survival is dependent upon the presence of water vapour in the atmosphere above 55% R.H. (Williams, 1970 b). This paper describes some effects of temperature and humidity on the water intake of G. colchici.

The method for obtaining continuous weight changes of these insects over a period of time and under known temperature and humidity conditions has been described previously (Williams, 1970 b). Briefly, the apparatus consisted of an electric microbalance, in which the weighing pan was suspended from a length of wire which ran down into an incubator through a hole in the top. The weighing pan was enclosed inside a chamber formed by a 250 cc staining jar, in which the temperature could be continuously monitored by a thermocouple and the humidity controlled by sulphuric acid solutions of varying strength (Solomon, 1951). For the present experiments the balance was used in three ways:

• (1) To obtain absolute weights.

• (2) To obtain weight changes of individual lice while the temperature was allowed to drop from 44·0 to 32·0 °C, the humidity being kept constant.

• (3) The weight changes of individual lice were monitored while the temperature was kept constant at one of six temperatures (44·0, 41·6, 39·2, 36·8, 34·4, 32·0 °C) various known humidities.

Most animals used in the following experiments had been made deficient in water by desiccation. This was accomplished by confining the animals in a Gooch crucible (Williams, 1970 a) with excess of food, at 0% R.H. and 35 °C. Under these conditions these insects lose water at approximately 1·6% of their initial body weight (W₀) per hour (Williams, 1970 b). Experimental animals were desiccated for periods of 10,14,16,18 or 20 h. The weight of these desiccated animals has been designated W_D in the text, and the weight of the animal after exposure to the test conditions W_t.

Two animals of each of the following groups were desiccated for 10,14,16,18 and 20 h and exposed to the temperature drop from 44 to 32 °C in 75 % R.H. : males ; non-fecund females ; females with a partially developed ovum ; females with a fully developed ovum. (For details of the technique for separating these females see Williams, 1970 a).

For any individual the temperature-drop treatment lasted from 55 to 65 min. The weight was recorded at 0·6 °C intervals. The rate of weight change per hour for each individual for every 0·6 °C drop was calculated. The results for a representative half of these experiments are shown in Fig. 1.

In the first instance it may be seen that many of the insects increased in weight during this treatment. Since the only source of weight gain available to the insect was the water vapour in the atmosphere, it must be concluded that these animals were removing water from their environment to replenish their artificially reduced water level.

It is also apparent that there were two extremes of active water uptake. At one extreme no water vapour at all was taken up (Fig. 1 : v, vii, viii and x) and at the other uptake was continuous (Fig. 1 : ii, iv, xvi, xvii, xviii, xix and xx). In the remaining cases uptake occurred only over a restricted part of the 44·0–32·0 °C temperature range (Fig. 1 : i, iii, vi, ix, xi, xii, xiii, xiv and xv).

It was also noted that, when it occurred, water-vapour uptake appeared to proceed at the same rate, 18–24% of the initial weight of the animal per hour, irrespective of the temperature.

Finally, attention was drawn to the observation that the various groups of female lice showed obvious differences in their patterns of water-vapour uptake. Non-fecund females, for example, did not take up any water vapour to replace their deficiency. Fully fecund females, on the other hand, took up water vapour continuously while being exposed to the temperature drop. Females developing an ovum characteristically took up water intermittently. Males, however, showed all three types of activity.

Having demonstrated that certain forms of G. colchici, if desiccated and exposed to a temperature drop in 75 % R.H., exhibit bursts of water-vapour uptake, the following experiment was designed to determined whether these bursts were temperature dependent.

Two animals of each of the following categories of animals were desiccated for 10 h and five for periods ranging from 14 to 20 h: males; non-fecund females; females with a developing ovum; females with a fully developed ovum. These individuals were prepared for exposure to each of the six temperatures in the 32–44 °C range. The weight changes of each individual were monitored for at least 1 h, or longer if weight changes continued past this period. In this latter case weighing was continued until the animal maintained a steady weight for 15 min. Typical curves for animals desiccated for 10 h are shown in Fig. 2(a), and for similar animals desiccated for 14–20 h in Fig. 2(b).

Water-vapour uptake was not continuous but occurred as one or more bursts of activity. Those animals with a water deficit of less than 10 % W₀ when exposed to 75 % R.H., initially remained at their desiccated weight for a period varying from 10 to 50 min, increased in weight for about 5 min, after which they maintained the new weight (Fig. 2 a). Rarely, a second period of activity occurred (Fig. 2(a), iii, iv).

Animals with a water deficit of more than 10% W₀ (Fig. 2a, b) usually began to take up water vapour immediately upon being placed in the environmental chamber. This activity continued uninterruptedly for periods ranging from 15 to 100 min, at which point the animals remained in equilibrium for at least 5 min. It was usual for _a second, and often a third, burst of activity to occur (Fig. 2(a), i-iv).

Analysis of the performance of the various animals at the different temperatures revealed that there was little variation in the rate of water-vapour uptake (Table 1) which was about 11% W₀/h.

Although neither the sex nor the reproductive state of the test animal nor the temperature affected the rate of uptake while it was taking place, it is fairly evident from examination of Fig. 2,(a, b) that both these factors did influence the extent to which uptake took place. It may be observed, for example, that animals with a water deficit of 15 to 25 % W₀ were not able to replace much of this deficiency in temperatures above 39·2 °C compared to animals held at 39·2 °C and 36·8 °C (Fig. 2 (b), i—iv). addition, it may be noted that non-fecund females desiccated for 10 h did not take up any water vapour at all (Fig. 2(a)), while all the other forms did.

To examine this more fully, the data for animals desiccated for 14–20 h have been analysed to bring out the differences between the various forms in the efficiency of water replacement at the different temperatures. The water-replacement efficiency (W.R.E.) for any individual has been defined as the amount of water taken up expressed as a percentage of the water lost during desiccation.

Thus, if the same amount of water was taken up as was lost during desiccation, the W.R.E. would be 100%. The result of this analysis is shown in Fig. 3. It may be seen that temperature had a similar effect on the W.R.E. in each of these adult lice. The average amount of water replaced was maximal in each case at 36·8 °C (62·1 %, males; 59·8 %, females with a developed ovum; 54·6%, females with a developing ovum; 19-9%, non-fecund females) and increasingly less water was replaced as the temperature increased or decreased on either side of this point. It was also apparent that the males and fecund females replaced a similar quantity of water at each temperature, whereas the non-fecund females usually replaced about one-third of this amount.

Since there was no significant difference between the males and fecund females in the amount of water replaced at each temperature, these results have been pooled for the purpose of statistical analysis (Table 2). This analysis shows that adult male and fecund female lice replaced most water (58-9 % of the water lost) at 36·8 °C. At 34·4 °C an average of 44·1% of the water lost was replaced, which was not significantly lower than that replaced at 36·8 °C measured by the Student t test. At 32·0 °C, however, 38·4 % of the water lost was replaced, which was a significant decrease on the value obtained at 36·8 °C (P > 0·005). Similarly, the amount of water replaced decreased significantly as the temperature increased above 36·8 °C (43·0 % of water replaced at 39·2 °C, P > 0·01 ; 25·4 % replaced at 41·4 °C, P > 0·001; 12·7 % replaced at 44·0 °C, P > 0·001).

Non-fecund female lice also replaced most water (19·9 %) at 36·8 °C. This decreased to 8·0 % at 32·0 °C, which was not significantly different. The amount of water replaced also decreased as the temperature increased from 36·8 to 44·0 °C. (14·6% at 39·2 °C; 9·0% at 41·4 °C; and 4·0% at 44·0 °C). Only at 44·0 °C was the decrease significant (P > 0 ·05).

Since non-fecund females replaced much less water than the males or fecund females, the data were re-analysed to obtain the W.R.E. of the various forms at different water, deficit levels, obtained as a result of varying the periods of desiccation. The result is shown in Fig. 4. It may be seen that the males did not reach their maximal W.R.E until they had lost 10 % of their body weight as water. Fecund females, however took up water vapour at maximal efficiency after having lost 5 % of their initial weight as water. Non-fecund females were able to lose 16% of their initial body weight as water before the uptake began to replace it, and even then the W.R.E. never approached that of males or fecund females. Having demonstrated that these differences in the efficiency of water replacement existed among the various adult forms of G. colchici, further series of similar experiments were performed to obtain information on the W.R.E. of immature forms at the various temperatures. Five or more third instar nymphs were desiccated for periods ranging from 14–20 h, and exposed to each of the six temperatures at 75 % R.H. Three individual third instar nymphs moulting to adults were treated similarly and exposed to each temperature. (Nymphs in the process of moulting could easily be distinguished from others by the absence of food in their gut, giving them a milky-white appearance. This stage normally lasted for between 48 and 72 h.) Five newly moulted adults were treated similarly and exposed to the six temperatures. (Newly moulted adults could be distinguished by their morphological similarity to adults, smaller size and much lighter colour. This stage would be distinguishable up to 72 h after the final moult.) The W.R.E. for each of these groups at the various temperatures is shown in Fig. 5 It may be seen that these immature forms had a different pattern of W.R.E. from that of adult lice. Lice undergoing a moult were unable to take up water vapour at all (Fig 5, ii) Some nymphs and newly moulted adults took up much more water (up to four times as much) than was lost due to desiccation. High and low temperatures in the 32 ·0–44 ·0 °C range did not appear to have any obvious depressant effect on the w R E. ; but there were always several animals at any one temperature which were unable to take up any water vapour at all.

All the above experiments were carried out in an environment of 75 % R.H. In order to examine the effect which various humidities had on the W.R.E., batches of ten males were removed from the incubator population, weighed, desiccated for 16 h, and then reweighed. (Males were selected for this experiment because of their consistency of response on previous occasions.) Seven of these batches were prepared, and one each was exposed to 100, 90, 75, 65, 55, 50 and 25 % R.H. at 35 ·0 °C. After 2 h the animals were reweighed. The mean W.R.E. for each humidity (except 55 % R.H., for which the individual results are plotted) is shown in Fig. 6.

It may be seen, first, that there was only a slight increase in the average amount of water replaced in humidities from 65 to 90 % R.H. (from 61 to 70 % of water replaced) although a substantial increase did occur in 100% R.H. (87% of water replaced), Secondly, no animal took up any water vapour in humidities of 50 % R.H. and below. There was a sharp transition humidity in the 50–60 % R.H. range above which the animals took up water vapour with maximal efficiency but below which there was no uptake at all. For this reason the results obtained at 55 % R.H. have been plotted individually, from which it may be seen that four of the ten males took up about 60 °/ of the water they had lost, five took up no water vapour at all, and one died.

Since none of the test animals took up water vapour after more than 2 h exposure to humid conditions, despite still being water deficient, it was possible that this could have been due to starvation.

To test this hypothesis ten fecund females were weighed, desiccated for 16 h without food and then exposed to 75 % R.H. at 35 °C for 2 h. A second group was treated similarly, except that they were allowed food during the desiccation treatment. Under these conditions it may be seen (Table 3) that females allowed food during the desiccation treatment had a mean W.R.E. of 38 ·2 %, and all took up water vapour. Animals starved during desiccation showed no uptake at all, and had a mean W.R.E. of–10 ·8 %.

It would be more useful to consider the basic properties of the water-uptake mechanism as revealed in these experiments before discussing its effects on the bionomics of Goniodes colchici.

Since Buxton (1930) demonstrated that starving Tenebrio molitor increased in weight in 90 % R.H. at both 23 and 30 °C, the phenomenon of water-vapour uptake has been shown to occur in so many insects, ticks, and mites as to make it an almost universal property of the arthropods (Mellanby, 1932; Kalmus, 1936; Ludwig, 1937; Lees, 1946, 1947; Edney, 1947; Govaerts & LeClercq, 1946; Browning, 1954; Beament, Noble-Nesbitt & Watson, 1964; Winston & Nelson, 1961 ; Kanungo, 1963 ; Wharton & Kanungo, 1962; Knulle, 1962, 1965, 1967; Belozerov & Seravin, 1960). In spite of these many studies, little is known of the mechanism of uptake. A brief review of present knowledge concerning the process is necessary before it is possible to consider the results of the present investigations.

There is much evidence that the uptake process is an active one, since it is not observed in arthropods which have been killed (Kalmus, 1936; Edney, 1947; Browning, 1954), anaesthetized (Browning, 1954), starved of oxygen (Lees, 1946; Browning, 1954; Kanungo, 1963), or poisoned (Lees, 1946; Belozerov & Servain, 1960). The threshold humidity above which the mechanism is able to operate is very variable, being as high as 90 % R.H. for some species, and as low as 45 % R.H. for others (Beament et al. 1964).

Buxton (1930) considered that the increase in weight which he observed in T. molitor was due to hygroscopic forces. Mellanby (1932) suggested that the site of uptake might be the tracheal endings, while Edney (1947) was of the opinion that water vapour was absorbed on to the cuticle, liquefied, and transported into the haemolymph. That the whole cuticle and not the tracheal endings is involved was demonstrated by Lees (1947) and Browning (1954), who both implicated the epidermis, as did Knulle (1965, 1967).

With the exception of Acarus siro (Knulle, 1965, 1967), the equilibrium weights for those arthropods studied do not vary in favourable temperature and humidity conditions. It was found possible for this equilibrium to change from day to day, corresponding to a change in form from larva to pre-pupa (Edney, 1947; Knulle, 1967).

From these experiments it has been inferred that the water-uptake mechanism in arthropods is uninfluenced by temperatures in the favourable range (Mellanby,1932; Edney, 1947; Wharton & Kanungo, 1962). However, in these cases continuous weighings were not attempted. It is clear in the case of G. colchici that temperature does affect the uptake process. The manner in which this occurs is not a simple one, since as has been demonstrated, the rate of water-vapour uptake is the same at all tempera.’ tures in the 32–44 °C range. However, since the uptake mechanism operates in bursts high and low temperatures in the 32–44 °C range impede water uptake by shortening the duration of these bursts, and temperatures in the middle of the range facilitate the process by allowing more frequent and longer bursts of activity. Therefore, although the rate of the uptake process is not influenced by temperatures in the 32–44 °C range what has been termed its efficiency, that is, the ability to take up a certain amount of water vapour in a standard time, is influenced by temperature.

Beament (1964) has discussed the theory of continuous and discontinuous secretory mechanisms (‘pumps’) and arrived at the conclusion that epidermal control of the cuticle, especially the proteinaceous component of it, would lend itself only to discontinuous activity. On other grounds he proposed that continuous ‘pumps’ could work if many factors, such as the asymmetric properties of isolated insect cuticles, lipid monolayer inversion, the electric-ion pump and various other things were taken into account. On the basis of the properties of the ‘pump’ as revealed in the present experiments the water-uptake mechanism is therefore probably located in the epidermal/integumental complex.

Accepting that the epidermal/integumental complex is responsible for water-vapour uptake in these insects, little is known of the factors controlling the system. There is some evidence that water levels in insects are controlled by hormones derived from the pars intercerebralis/corpus allatum complex, and the corpus cardiacum, the former tending to elevate water levels, and the latter to depress them (Pfeiffer, 1945; Koidsumi, 1952; Altmann, 1956a, b; Nayar, 1956; Nuñez, 1956). In addition, Clarke & Langley (1962) have shown that in Locusta migratoria the liberation of the hormones derived from the brain/corpus allatum complex is under the control of nervous impulses arising from movements of the foregut during the course of feeding. The typical feeding behaviour of ischnoceran lice, in continually cropping feathers, would ensure a constant supply of corpora allata hormones necessary to elevate the water level. Starvation would remove this stimulus, resulting in the cessation of hormone release and a consequent inability to take up water vapour. This seems to be the most probable reason why moulting lice are unable to take up water vapour, since feeding ceases during moulting. This would also explain why desiccated adult lice are rarely able to take up water vapour after 2 h exposure to humid conditions, despite still being water-deficient, since at this time the effects of starvation are probably beginning.

These results therefore are consistent with the hypothesis that water levels in G. colchici are determined by the level of hormones secreted by certain of the neurosecretory organs, and that the principal mechanism in water level adjustment is the epidermal/integumental complex in its capacity to take up water vapour actively.

It remains to consider how the secretory mechanism as revealed in these investigations affects the biology of G. colchici.

It has been shown that adult male lice are only able to take up water vapour in humidities above 50% R.H., and that the threshold for any individual is sharply defined in the 50–60 % R.H. range. In a previous paper (Williams, 1970b) it was shown that the threshold humidity for survival of all adult G. colchici follows an identical attern. This strongly suggests that G. colchici relies upon the active uptake of water P^a_our f_or the maintenance of its water balance, and thus for survival.

For most of the time the bulk of the water being lost by an individual louse will be that due to transpiration, which occurs at the rate of approximately 1 ·6% of the body weight per hour (Williams, 1970b). The water-vapour uptake mechanism, at maximal efficiency, can take up water at the rate of 11 % of the body weight per hour. Thus transpiration losses can easily be replaced when the water-vapour uptake mechanism is working well below maximum capacity.

At other times certain lice will lose more water than that due simply to transpiration. Animals undergoing a moult, for example, will have a compound water loss, including that lost during the shedding of the cuticle in addition to transpiration losses. Since, for the duration of moulting, the uptake process does not function, the insect will emerge from the moult with a greatly reduced water level. Even were the animal not to be water-deficient, more water would be required for growth in the initial post-moulting phase than normal. In the immediate post-moulting stage, therefore, it is imperative that the uptake mechanism be working efficiently. (The demonstration that some newly moulted adults and third instar nymphs gained up to four times the amount of water lost during the desiccation treatment, while adults never did, points up the validity of these suggestions.)

It also emerged from these experiments, however, that at any one of the six temperatures examined in the 32–44 °C range there were always a substantial number of newly moulted animals unable to take up any water vapour at all. This would suggest that for any individual louse in the post-moulting stage there exists, as has been demonstrated for adults, a restricted temperature range over which the uptake of water vapour is most efficient. It would also appear that for any of these individuals this could be anywhere in the 32·0–44 ·0 °C range studied. Therefore, if all these animals were to be confined at one temperature, large numbers of them, being unable to replenish the water loss incurred during moulting, would die. Since the ischnoceran lice are usually artificially cultured at one temperature, this would account for the large post-moulting mortalities noted by previous workers on lice bionomics (Wilson, 1939; Ash, 1960). Under natural conditions individual lice would be able to select the desired temperature by migrations along the rachis of the feather in order to overcome this problem (Williams, 1970a).

It may also be postulated that fecund female lice are also subject to a greater than normal rate of water loss. It has been shown that at 35 ·0 °C, G. colchici lay eggs at the rate of about one each day (Williams, 1970 a). Since a fully formed ovum occupies a considerable volume of the abdomen of these animals, and since the eggs probably contain a large percentage of water, the loss of water from the female at egg-laying must be considerable. In order that eggs should continue to be laid at this rate, the water being lost must continually be replaced. (This suggestion is supported by the observation that non-fecund lice were able to lose up to 16% of their body weight as water before commencing to replace it, whereas fecund females took up water vapour at maximal efficiency after having lost as little as 5 % of their initial weight.)

It has been demonstrated that adult lice secrete most water vapour at 36 ·8 °C. If the hypothesis that egg production relies upon the efficient operation of the Water, uptake process is correct, eggs should be laid in the largest numbers at temperatures close to 36 ·8 °C, and decrease as the temperature rises and falls about this point This is indeed the situation, both in the case of G. colchici (Williams, 1970a), and of other ischnoceran lice which have been studied (Mathysse, 1944; Arora & Chopra 1957; Murray, 1957, 1960).

In addition, since very little increase in the efficiency of water-vapour uptake occurs with increasing humidities from 55 % to 90 % R.H., it may be inferred on the basis of the above hypothesis that increasing humidities within this range will not result in any increase in egg production. This has been demonstrated for G. colchici (Williams 1970 –a, b)

Furthermore, it is possible to explain the phenomenon of egg-resorption in these insects, described in an earlier paper (Williams, 1970a), if the production of ova is dependent upon humidity and temperature operating through the active uptake of water vapour. Thus, a mature female louse held at temperatures in the vicinity of 36 ·8 °C would be able to take up sufficient water vapour to produce eggs at the rate of one each day. Should this louse be removed from 36 ·8 °C to a higher or lower temperature, the quantity of water being taken up would decrease and would ultimately be insufficient to allow the maturation of any ova being developed. In this circumstance it is probable that the egg would be resorbed. If this is correct, egg resorption in G. colchici should occur mainly when these lice are switched from temperatures close to 36 ·8 °C to temperatures rising to 40 ·0 °C, or decreasing to 30 ·0 °C. This has been demonstrated (Williams, 1970a).

The author wishes to express his sincere thanks to Dr K U. Clarke, who acted as supervisor during the course of this investigation. He would also like to thank Professor E. J. W. Barrington, for making available the facilities of the Department of Zoology, Nottingham University, and the Science Research Council, for advancing a substantial grant allowing the purchase ot the micro-balance, without which this study would not have been possible. The author is also indebted to R.S Balter,Esq., for the loan of much equipmeant and good will. The study was carried out under a studentship provided by the Science Research Counsil, to whom the author is most grateful.