By the methods described in this paper it has been shown that:

  1. A clasping action can be produced in male frogs at all times of the year.

  2. This reaction is peculiar to the male.

  3. The course of the decline and onset of the hypertrophy of the brachial musculature up to a maximum at the breeding period can be followed.

  4. This hypertrophy appears to follow fairly closely the changes occurring in the interstitial tissue of the testis.

The sexual behaviour of frogs at the spawning period is a well-known phenomenon and has been described by several observers. The clasping reflex in particular has attracted attention, and numerous investigations have been made on the subject, special attention being paid to the factors governing the manifestation of the embrace and the mechanism whereby the male can distinguish the female (Spallanzani, 1784; Goltz, 1865; Tarchanoff, 1887; Meisenheimer, 1911 ; Nussbaum, 1909; Savage, 1934). Several of these workers showed the presence of an inhibitory centre in the midbrain, stimulation of which caused the cessation of the embrace and the disappearance of sexual activity for a short time. The clasping reflex has been shown to be a true secondary sexual character, and is dependent on the presence of the testicular hormone (Allen, 1932; Aron, 1926; Smith & Schuster, 1912). In addition, the reflex has been found to be dependent on the presence of an anterior pituitary hormone (Rugh, 1934; Bugbee et al. 1931; Adams, 1931, etc.).

In the course of seasonal observations on the frog (Rana temporaria) the attention of the writer was arrested by the remarkable persistence of the clasp reflex in the male despite strong nociceptive stimulation. Pl. I, figs. 1 and 2 are photographs of preparations made to illustrate this. The male was pithed and the head almost severed from the body (fig. 1), but there was no interruption of the embrace. In fig. 2 a male is shown with the hinder part of the body also severed, leaving only the pectoral girdle and the upper four or five vertebrae attached to the female. After this operation the reflex persisted for about half an hour, by which time the nervous excitability of the preparation was practically destroyed. Toads (Bufo vulgaris) were also treated in the same way and similar results obtained.

Text-fig. 1.

Sketch of apparatus for recording the intensity of the clasp reflex. (For explanation see text.)

Text-fig. 1.

Sketch of apparatus for recording the intensity of the clasp reflex. (For explanation see text.)

Text-fig. 2.

Tracing showing lack of response of female frog to electrical stimulation.

Text-fig. 2.

Tracing showing lack of response of female frog to electrical stimulation.

That the midbrain is the site of an inhibitory centre was also confirmed, for it was found that gentle stimulation of this region with the point of a needle caused immediate separation in both frogs and toads. It was also noticed that, as Tarchanoff stated, there is no visible grave disturbance of the male, the forelimbs simply being unclasped, and the frog usually slips sideways from the back of the female. Inhibi-tion is complete for a short period, but the sex instinct slowly returns and embracing will again take place in about half an hour.

It was observed that a clasping action could be induced in male frogs after the normal breeding season by stimulation with a faradic current. Any point of stimulation is effective provided the stimulus can be conducted to the brachial muscles. Thus a response is obtained if the electrodes are placed on the skin of the ventral surface between the arms, as well as by stimulation of the brachial nerves or the central nervous system. A method of recording this induced clasp has been devised and is described in the next section.

II. METHOD

The following apparatus was designed to obtain a graphic record of the clasp reflex produced in both frogs and toads by stimulation with a faradic current.

The principle of the method can be seen from Text-fig. 1.

A male frog is placed on a piece of thin rubber tubing (A) (a long narrow balloon with the closed end cut off) stretched between two short pieces of glass tubing, one of which is closed, the other piece being connected by rubber tubing with a wider glass tube (B). The rubber tube A, the connecting tubing and the glass tube B are filled with water, the level in B being maintained a little higher than A to distend the balloon by a slight positive pressure. A long needle is floated vertically in the tube B by means of a suitable piece of cork, and its upper end is connected to a long suitably counterpoised lever as shown. This records any change in the water-level on a revolving drum. A pair of electrodes (E) are led off from the secondary coil of a Rhumkorff coil, the primary circuit of which includes a key, two 1·5 V. dry cells and an electromagnetic signal (S). The electrodes are placed in contact with the part of the frog which is to be stimulated and the key closed. If the clasp reflex is evoked the forelimbs grip the rubber tubing (A) and the resulting increase in pressure causes the lever to fall. By arranging the signal lever and the recording lever exactly one above the other the exact point of stimulation is shown. The time signal is given in seconds.

An important point is the diameter of the rubber tubing which the frog is induced to embrace. During the breeding season it was found that an approximate diameter of 1 in. was quite satisfactory for toads, but with frogs, and more particularly after the end of the breeding Reason, this was found to be too large, so that a rubber balloon of in. diameter was substituted, as stated above, with satisfactory results.

III. EXPERIMENTAL RESULTS AND DISCUSSION

The typical form of the record obtained can be seen from Text-figs. 3–6. The type of tracing was practically the same under all conditions of stimulation, although when the central nervous system was stimulated it was usually noticed that there was appreciable “after-discharge”. That is to say, when the skin of the breast region is the point of stimulation the clasping action ends abruptly with the cessation of the stimulus, but if the electrodes are applied to the cut end of the spinal cord in a decapitated animal, then the clasp is abolished more slowly. The latter type of tracing is shown in Text-fig. 5, where it can be seen that the time between the cessation of the stimulus and the return of the lever to rest represents an interval of approximately 12 sec. In the other figures which were obtained by stimulation of the skin of the breast the end of the reaction practically coincides with the removal of the stimulus.

Text-fig. 3.

Series of records showing the independence of the induced clasp and the central nervous system. (i) Response of normal frog to breast stimulation (ii) Dorsal spinal nerves cut. (iii) and (iv) Brachial nerves cut. (v) Stimulation of the spinal cord

Text-fig. 3.

Series of records showing the independence of the induced clasp and the central nervous system. (i) Response of normal frog to breast stimulation (ii) Dorsal spinal nerves cut. (iii) and (iv) Brachial nerves cut. (v) Stimulation of the spinal cord

Text-fig. 4.

Record of clasp induced by stimulation of the breast region of a male frog, 31/7/34.

Text-fig. 4.

Record of clasp induced by stimulation of the breast region of a male frog, 31/7/34.

Text-fig. 5.

Record of clasp induced by stimulating the brain of a male frog, 11/10/34.

Text-fig. 5.

Record of clasp induced by stimulating the brain of a male frog, 11/10/34.

Text-fig. 6.

Record of clasp induced by stimulation of the breast region of a male frog, 11/3/36

Text-fig. 6.

Record of clasp induced by stimulation of the breast region of a male frog, 11/3/36

It was at first thought that measurement of the latent period for the incidence of the induced clasp might be of value, but with this type of apparatus it did not prove possible. The latent period of the apparatus itself is much greater than that of the preparation, so that measurements, of the degree of accuracy possible with the relatively slow drum speed, would be worthless.

In order to see whether this artificial clasp was peculiar to the male, control experiments were carried out on female frogs and toads by similarly stimulating them. That there was no sign of a clasping action is shown by the straight line recorded in Text-fig. 2 which shows a typical tracing. Even to the eye the nature of the response was different, the forelimbs of the female being extended rather than flexed, while in some cases they were raised above the head in an attempt to dislodge the electrodes when the central nervous system was stimulated.

As regards the clasp obtained by stimulating the skin of the breast, there would appear to be two possible ways by which this could take place: (a) the current is conducted through the intervening tissues and acts directly on the brachial muscles; (b) the reaction might be due to stimulation of sensory endings in the skin which transmit afferent impulses to reflex centres in the cord.

Text-fig. 3 shows the record of an experiment designed to investigate the mode of action. All the tracings, except one, are the result of stimulating the skin between the forelegs, the first one being on an intact frog. The animal was then pithed and the dorsal spinal nerves severed. As a clasp was still obtained the brachial nerve of both sides was cut, and it is evident that a strong reaction was still obtainable. The final part of the figure is the record of stimulation of the spinal cord, and the lack of response shows that connexion with the central nervous system had been abolished. This experiment definitely shows that the electrical stimulus is spreading from the skin and acting directly on the muscles, the sensory nerves and the central nervous system apparently not being necessary for the appearance of the reaction.

Text-figs. 4–6 show the increase in intensity of the induced clasp which takes place during the year. From July, when the first records with the apparatus in its final form were obtained, to March when normal pairing was occurring, a progressive increase in the height of the curves can be seen. This increase is not uniform, being more marked at certain periods than others. Aron (1926) states that the musculature of the forelimbs in the male frog is greatly hypertrophied at the breeding season and that the difference between male and female is striking. The graphic records shown in the figures confirm this statement, and in addition they show that the hypertrophy is taking place gradually during the winter. Table I shows the average heights of the curves obtained for samples of frogs at different times of the year. Care was taken to render the different records obtained as comparable as possible. The secondary coil in the electrode circuit was kept at a standard distance from the primary (namely, 7 cm.), and the frogs used ranged in weight from 20 to 25 g. Apart from this, however, there are other inevitable experimental variations, for instance, the friction of the writing point on the drum will vary slightly from one experiment to the next. Also the intensity of the clasp produced varies to some extent according to the manner in which the frog grips the balloon, but this is mitigated as far as possible by taking from four to five records on each specimen. Each sample of frogs usually contained about six males, so that it is considered that the data in Table I gives a fairly true picture of the progress of the hypertrophy of the brachial muscles.

Table I.

Showing the variation in intensity of the induced clasp

Showing the variation in intensity of the induced clasp
Showing the variation in intensity of the induced clasp

Table I is reproduced graphically in Text-fig. 7, where the curve given by Aron (1926) for the seasonal development of the interstitial tissue in the frog testis has also been included. As regards the artificial clasp it is apparent that the first significant increase in intensity occurs in October and November. In December it is still at approximately the same level, but thereafter follows a marked rise up to the breeding season. This sequence of events agrees very well with Aron’s observations on the interstitial tissue, for it is apparent that his curve also shows two periods of rapid hypertrophy, from July to October, and from December to February. From this figure it would appear that the muscular hypertrophy follows some time after the beginning of the increase in the interstitial tissue and this may indicate a causal relationship between the two. If the interstitial tissue of the testis is responsible for the production of the sex hormone, then it would be expected that the development of a secondary sexual character such as the brachial muscles would be consequent on the hypertrophy of the former. Aron shows that the development of the nuptial pads is correlated with the interstitial tissue in a very similar way.

Text-fig. 7.

Graph showing the seasonal hypertrophy of the brachial musculature in the male frog. ×— ▪ — · showing the seasonal variation in the artificial clasp. ○— ○ — ○ curve of the development of the interstitial tissue of the testis. (After Aron, 1926)

Text-fig. 7.

Graph showing the seasonal hypertrophy of the brachial musculature in the male frog. ×— ▪ — · showing the seasonal variation in the artificial clasp. ○— ○ — ○ curve of the development of the interstitial tissue of the testis. (After Aron, 1926)

It is necessary to consider briefly the reasons for the difference in response to stimulation between males and females noted above. Since the clasp can be produced equally well in the male by stimulating the central nervous system, the brachial nerves or the muscles themselves with a strong faradic current, it is apparent that the whole of the brachial musculature will be involved. As the phenomenon is peculiar to the male there must be a sexual difference in the musculature of the forelimbs. Aron’s observation that there is a striking difference in this respect between males and females at the breeding season has already been quoted, but this investigation shows that the difference persists throughout the year, although it is less marked during the summer. When the male frog is stimulated the flexor and the extensor muscles will be operating antagonistically, but as a clasping action results the flexors must be sufficiently strong to overcome the extensors. The change in the intensity of the artificial clasp during the year is thus probably due to alterations in the development of the flexor muscles alone. In the female frog, however, the extensor muscles appear to be stronger than the flexor muscles at all times, so that the response to stimulation is different. It may be that the matter is not quite so simple as this, for stimulation of the female does not necessarily involve rigid extension of the forelimbs, as purposive movements to remove the stimulus are sometimes made. It is not, however, suggested that the normal clasp reflex of a mating frog involves the contraction of both flexor and extensor muscles. In this case the reciprocal innervation of the antagonistic muscles will probably lead to inhibition of tone in the extensor muscles when the flexors are contracting. It seems very doubtful whether this reciprocal inhibition could be possible in the experimentally induced clasp, owing to the mode of stimulation, but it may be a possibility. It would be of interest to study the extent of the hypertrophy of the brachial muscles by this method under the influence of various treatments. For instance, does any hypertrophy take place when amplexus is induced by anterior pituitary injections? Again, if this muscular development is directly controlled by a testis hormone, it might be possible to obtain clasping action in a castrated female into which testicular implantations had been made.

The author wishes to express his appreciation of the invaluable assistance and criticism given by Prof. J. H. Orton and Mr S. T. Burfield throughout the course of this work.

Adams
,
A E
(
1931
).
Proc Soc exp. Biol., N.Y
.,
28
,
677
.
Aiien
,
E
(
1932
)
Sex and Internal Secretions
Baltimore
.
Aron
,
M.
(
1926
)
Arch. Biol, Paris
,
36
,
1
97
.
Bugbee
,
E P
,
Simond
,
A E
&
Grimes
,
H. M.
(
1931
).
Endocrinology,
25
,
41
Goltz
(
1865
)
Cent dtsch Med. Wiss
1865
6
.
Musenheimer
,
J.
(
1911
).
Zool Anz.
38
,
53
.
Nussbaum
,
M
(
1909
).
Pflug Arch. ges. Physiol
126
,
519
Ruch
,
R.
(
1934
)
Biol. Bull. Wood’s Hole,
66
,
22
29
Savage
,
M. R
(
1934
).
Proc zool. Soc Lond.
pp.
55
70
.
Smith
,
G. E
&
Schuster
,
E.
(
1912
).
Quart. J. micr Sci.
17
,
439
469
.
Spallanzani
,
L.
(
1784
).
Dissertations, 2
.
London
.
Tarchanoff
,
J. R.
(
1887
).
Pflug. Arch, ges Physiol.
40
,
330
.
Plate 1

Fig 1. Showing the persistence of the clasping reflex in a decapitated male frog

Fig. 2. Male frog with head and also the posterior part of the body cut off ; the clasping reflex is still maintained

Plate 1

Fig 1. Showing the persistence of the clasping reflex in a decapitated male frog

Fig. 2. Male frog with head and also the posterior part of the body cut off ; the clasping reflex is still maintained