A QUANTITATIVE STUDY OF THE BEHAVIOUR OF THE

MALE MORMONIELLA VITRIPENNIS (WALKER)

(HYMENOPTERA, PTEROMALIDAE) TOWARDS TWO
CONSTANT STIMULUS-SITUATIONS
by
ROBERT BARRASS 1)
(Department of Zoology, The University, Nottingham, England.)
(With 13 Figs)
(Rec. 20-III-1961)

INTRODUCTION
Much of the evidence for the concepts of reaction specific energy (LORENZ
1950) and specific action potentiality (HINDE 1954b; LoRE?Tz 1956) was
based on studies of the waning in an animal's responsiveness to a constant
stimulus-situation. This response waning has been observed in insects (TIN-
BERGEi\T Ztail 1943) and in birds (TINBERGEN & PERDECK 1950; HINDE
1954b). The nervous systems of these animals are so very different however
and there is no suggestion that the resemblances are manifestations of the
same process.
The use of a model, that is of an inadequate stimulus (THORPE 1956), in
these studies, may have meant that only a part of the response could be per-
formed. The present paper reports an attempt to determine whether or not
this waning still occurs when the complete behaviour pattern is performed and
it is also concerned with changes in the responsiveness of the male M. vitri-
pennis towards two distinct stimulus-situations. One of these is the receptive
and the other the non-receptive female M. vitripennis. The behaviour
considered is the courtship.
This courtship behaviour has been described (BARRASS ig6oa) but may be
briefly summarised. The male chases the female, mounts and then moves
until its head is above and between the raised antennae of the female. In this
courtship position, movements of the male's head, antennae and wings occur

1) The work included in this paper formed part of a Ph.D. thesis (1956) at Nottingham
University. I wish to acknowledge the kindness of Professor E. J. W. BARRINGTON and
to thank Dr K. U. CLARKE, who supervised, for his encouragement and helpful criticism.
I also thank Professor E. B. EDNEYand Dr W. H. THORPEfor reading the manuscript
of the present paper and for their comments. Dr A. J. WATSONkindly allowed me to see
the proofs of his paper (1961) prior to publication. The French summary was prepared by
Professor B. H. RASMUSSEN. The work was made possible by a post-graduate award
from the Lancashire Education Committee.
 289

and of the tarsi of the prothoracic limbs. A movement in which the male's
head is raised and then lowered in relation to the female's antennae is called
a h e a d m o v e m e n t. Several of these movements are performed in suc-
cession as a h e a d s e r i e s and a courtship usually includes a number of
head series with an interval of a few seconds between them. A receptive
female lowers its antennae as the male raises its head at the start of a head
series and this releases the copulation attempt. Following copulation the male
moves forwards again to the courtship position and courtship movements are
repeated in a p o s t - c o p u 1 a t o r y courtship. The male then dismounts.
M. vitripennis is a pupal parasite of cyclorrhaphous Diptera. The syno-
nomy of this insect has been considered by IVHITING ( y58) ; other names
include Nasonia brevicornis Ashmead and Nasonia vitripennis (Walker).
Puparia of Musca doyvcestica L. were used as host.

METHOD

The method of dual quantification (LOREVZ y5o) is applied; for which

it is necessary to standardise both the experimental animal and the stimulus-
situation presented to it.
Changes, due to age, in the duration of courtship and in the number of
movements performed (BARRASS, ig6ob) were avoided by using males which
had been raised in isolation from the pupal stage onwards and which were
in the first day of imaginal life. The males were not fed.
A virgin female almost invariably allows copulation in its first courtship
(CousiN, 1933). The behaviour of a female is, however, profoundly affected
as a result of this first courtship, and females taken from a stock culture of
both sexes are non-receptive. Females in their first day of life were not fed.
Older females were fed on honey and host blood. The receptive females were
used on the first or second day of imaginal life, but when very large numbers
were required they were sometimes older. The non-receptive females were
not standardised with regard to age.
The distinction between r e c e p t i v e and n o n - r e c e p t i v e females
makes possible a study of a male's behaviour towards two constant stimulus-
situations, one of which (the receptive female) releases the complete be-
haviour pattern, and the other (the non-receptive female) only releases a part
of the pattern.
Several measures of responsiveness are used in this analysis: -
i. The duration of courtship.
2. The number of head series in a courtship.
3. The number of head movements in a courtship.
290

4. The number of head movements per head series (series cut short by
a copulation attempt are not included).

5. The number of courtships performed by a male when a large number of

females were presented to it at short time intervals over a long period
of time.

When no courtship movements occurred during an observation a zero was

recorded which could refer to several different types of behaviour, namely,
a male mounting and then dismounting without courtship, or, a male starting
to mount but not doing so, or, a one minute observation without an attempt
at mounting.
All observations were made in a clean glass observation cell ( r mm. in
diameter and 4 mm. deep), which had a glass roof but no floor. A binocular
microscope was used and a lamp giving a light intensity of 500 to 550 foot
candles. The temperature of the cell was the same as room temperature,
21 ± 3° C. (the cell was opened frequently for the introduction of new
females) and humidity was not controlled.

RESULTS
(a) THE MALE'S COURTSHIP OF NON-RECEPTIVE FEMALES

When a male was observed with a non-receptive female, changes in its

behaviour were noted at successive contacts. The first contact was typically
followed by mounting and a number of courtship movements were performed
(X = 6.95 head series, 34.14 head movements, 4.95 head movements per head
series: - means of y observations). Fewer head series occurred in suc-
ceeding courtships (Fig. r ) and then there were many contacts which included
only the earlier parts of the behaviour pattern; the male sometimes mounted
and then dismounted at once, or started to mount without doing so, or turned
and chased without attempting to mount, although it was often very close to
the female or even touching it. These changes in a male's responsiveness
towards a constantly present female are not easily analysed because the time
interval between contacts is very variable; this is especially true later in an
observation when the female may be inactive. To overcome this difficulty
the female was removed at the end of the courtship and another female intro-
duced for each successive observation. The time from the end of one
courtship to the start of the next observation could then be fixed, giving a
fairly accurate control of the time interval between contacts. When this time
interval was very short, the male's behaviour towards successive females was
virtually towards a constantly present stimulus-situation.
 291

Fig. i. The behaviour of three males. Each male was observed with a non-receptive
female and the first 30 contacts are recorded. The length of the vertical line shows the
parts of the courtship behaviour observed at each contact.
'
Experiment One.
Each male was placed with a female and courtship was observed. Sub-
sequent observations followed at fixed time intervals until the male had been
presented with ten non-receptive females in succession. The fixed intervals
were 0 min, i min, 3 min, 10 min, and 30 min. 4o males were used at the
o min. interval and io males at each of the other time intervals.
At the half-minute interval, the male spent most time on the female, per-
formed most head movements and most head series, during the first obser-
 292

Fig. 3. Effect of length of time between

" courtship upon the number of head move-
SUCCESSIVEOBSERVATIONS ments in successive courtships. M = a mi-
Fig. 2. Effect of length of time between nute.
courtship upon the duration of courtship.
M = a minute.

Fig. 4. Effect of length of time between Fig. 5. The number of head movements per
courtships upon the number of head series head series in successive courtships occurring
in successive courtships. M = a minute. at different time intervals. M = a minute.
 293

vation. There was then a progressive decrease in the amount of time spent
upon the female (Fig. 2), in the number of head movements (Fig. 3) and in
the number of head series (Fig. 4) in the later observations. These changes
were due to two factors, firstly to a decreasing tendency to court the female
in the later observations and secondly to a decrease in the duration of
courtship and the number of movements performed when courtship did occur.
At the same time, there was an increase in the mean number of head move-
ments per head series in the successive observations (Fig. 5), 4.7 in courtship
one and varying between 6.04 and 6.56 in courtships 4 to 10.
With a longer time interval between observations the time spent on the
female (Fig. 2), the total number of head movements (Fig. 3) and the num-
ber of head series (Fig. 4) are all higher than at the half-minute interval.
Whatever the time interval, however, there was an increase in the number
of head movements per head series in the later courtships (Fig. 5).
At the shorter time intervals there was a progressive restriction of the
courtship movements to the earlier parts of the behaviour pattern. There
was, first of all, a decrease in the number of courtship movements performed
and then a decreasing tendency to mount, but in all observations the male
turned towards the female and chased. When the observations were at 30
minute or io minute intervals, the male chased the female and almost always
mounted and courted.

Experiment Two.
Each male was allowed one courtship of a non-receptive female and then
four courtships at each of the time intervals: 30, 10, 3, 1, %, 1, 3, io and 30
minutes. Thirteen observations of this kind were started and from these
the results of eight observations were selected. Selection of this eight
demanded that the male involved should court the first nine females.
The mean behaviour of these eight males is summarised in table I. The 32
observations at each time interval were averaged to give the arithmetic means
shown. As the time interval between observations decreased there was a
decrease in the time spent upon the female, in the number of head movements
and in the number of head series but an increase in the number of head
movements per head series. As the time between observations increased these
changes in the male's behaviour were reversed. At the end of this experiment
the male was taking longer to perform a larger number of head movements
in approximately the same number of head series. Consideration of the
numbers of head movements per head series reveals that 'recovery' was not
complete at the end of the experiment and also that there is a time lag when
294

TABLE I
Duration of courtship (rnean of 8 6 1 : 32 obs.) and number of courtship
zuovenaents when the interval between courtships was progressively decreased
and then progressively increased.
Duration Number of Number of Head
of head head movements
courtshiP movements series per series

the results of the second half of the experiment are compared with those of
the first half.
Further information regarding the effect of courtship upon the male was
obtained by a different experimental procedure in which the male's behaviour
was studied in terms of the courtships performed and the distribution of these
courtships in time, when a non-receptive female was present throughout a
period of several hours duration.

Experiment Three.
The stimulus-situation was a non-receptive female moving about the cell.
To investigate a male's responsiveness to this situation over a long period of
time many females had to be used since a single female soon became inactive
and consequently the stimulus-situation was altered. One male was observed
with a succession of non-receptive females. It was possible to remove one
female from the cell and to replace it by another in i5 sec. but over a 3 to 5
hour period this time varied between 15 and 30 sec. In this way the
behaviour of each of six males was observed for several hours towards a
stimulus-situation which was virtually constantly present.
Some facts concerning the behaviour of these six males are shown in table
II and a more detailed record of the behaviour of two of them (Male 5 and
Male 6) is given in Figs 6 and 7. The interval between courtships was fairly
Constant but there was considerable variation in the male's response and there-
fore in the duration of an observation; the number of observations in one
hour varied as a result.
 295

In the first observations each male showed a decreasing tendency to per-

form the complete courtship pattern and when courtship did take place fewer
head series occurred than in the first courtship. In later observations
courtships were infrequent with intervals of almost one hour in the case of
some males (Males 1, 2 and 4) and when eventually a courtship did take
place, a large number of head series sometimes occurred, perhaps more than in
the male's first courtship. During these long intervals between actual courtship
the earlier parts of the courtship pattern were seen in almost every observa-
tion. The male turned and chased as the female moved, sometimes placing its

Fig. 6. A record of the behaviour of male 5 (Experiment 3; Table II) towards 201
successive non-receptive females over a period of 3 hours 42 minutes. The length of each
vertical line shows the parts of the courtship behaviour observed with each female. A
black dot on the base line shows that the male did not attempt to mount during a one
minute observation.
296

prothoracic limbs upon the female only to move away again without mounting.
Less often, the male mounted but then dismounted without moving to the
courtship position, resuming its turning and chasing movements almost at
once.

Fig. 7. A record of the behaviour of male 6 (Experiment 3; Table II) towards

209 successive non-receptive females over a period of 4 hours. Diagrammatic representa-
tion as for figure 6.

(b) THE MALE'S COURTSHIP OF RECEPTIVE FEMALES

Experiment Four.
The stimulus situation is a receptive female moving about the cell. Each
of six males was with a succession of receptive females. The observations
 297

were made as in the previous experiment so that, once again, the female was
virtually constantly present.
The results in table III refer to this experiment and a more detailed
record of the behaviour of two of the males (Male 4 and Male 5) is given
in Figs 8 and 9. Copulation did not take place in every courtship; sometimes

TABLE II
Results of f observations of six separate males; each with a succession of non-
receptive females over a period of several hours.
Duration Time Number Number Number Number Number
of upon of of of of Head of Head
Series Female Females Courtships Copulations Series Movements

TABLE TTT
Results of observations of six separate males; each with a succession of
receptive females over a period of several hours.
Duration Time Number Number Number Number Number
upon of of of of Head of Head
Female Females Courtships Copulations Series Movements

the male did not attempt copulation although the female was receptive and in
other observations the male courted without the female adopting a receptive
attitude. This lack of response was probably due, in most cases, to the male's
courtship being too brief, (in 99 observations of a male's first courtship of
receptive females, 36% of females responded at the first head series, 73 Jo
at the first or second series and 90% by the beginning of the third head
series), but sometimes the female did not respond even when the male's
courtship was prolonged.
In this experiment each male courted many females and spent a consider-
able part of the total observation period actually upon the female. With the
exception of male six, more courtships occurred in the first hour of each
298

observation than in the next hours. The most striking point is that one male
is capable of so great a number of copulations in quite a short time, for
example, male four (Fig. 8) courted 157 females in a period of four hours
twenty-four minutes and copulated with 154 of them.
All males turned and chased the females throughout these observations
although in some cases mounting and courtship did not follow. The courtship

Fig. 8. A record of the behaviour of male 4 (Experiment 4; Table III) towards 208
successive receptive females over a period of 4 hours 24 minutes. The length of the
vertical line shows the parts of the courtship behaviour observed with each female and a
horizontal line through it shows at what point mating occurred. A dot above the
horizontal line shows a courtship which did not include a post-copulatory courtship. A
dot on the base line shows that the male did not attempt to mount during a one minute
observation. A cross below the base line shows a courtship in which the female became
receptive without the male attempting to mate.
 299

of each male was typical at first but after many copulations had occurred
some of the later courtships terminated with copulation (there was no post-
copulatory courtship). These courtships are indicated in Figs 8 and 9 by
means of a dot immediately above the observations concerned. When, how-
ever, a post-copulatory courtship did occur, a second copulation followed on

Fig. 9. A record of the behaviour of male 5 (Experiment 4; § "Fable III) towards 214
successive receptive females over a period of 4 hours 52 minutes. Diagrammatic
representation as in figure 8.

y occasions and two males copulated three times in a single courtship (male
i with the seventh female and male 4 with the i85th female). A second
copulation has never been observed in a male's first courtship.
In the later observations a male sometimes evoked a response in the female
but then remained in the courtship position (x below line in Figs 8 and 9).
300

When eventually the female raised its antennae, courtship was resumed and
if the female was again receptive copulation invariably followed.
Many more courtships occurred in these observations with receptive fema-
les than with non-receptive females and, apart from the types of behaviour
mentioned above, there was no restriction of courtship to the earlier parts of
the pattern. Two exceptions to this statement were males 3 and 6 which did
not court during long periods although the first parts of the pattern were
observed. Both of these males courted "non-receptive" females just before
the periods without courtship and this perhaps explains the change of be-
haviour.

(c) A COMPARISON OF THE EFFECTS OF COURTSHIP OF NON-

RECEPTIVE AND RECEPTIVE FEMALES UPON THE MALE'S
BEHAVIOUR
The effects of one presentation of a stimulus upon an animal are
measurable by comparison with the response to a second presentation of the
same stimulus immediately afterwards. It can be argued (see also HINDE
1954b) that the difference between the first and second response is a result
of the first response and a measure of its effects. This method was used to
study the effects of courtship upon the male's subsequent behaviour and, by
using several time intervals, the recovery from these effects.

Experiment Five.
Each male courted a non-receptive female and then after a certain time
(;/z min., i min., 3 min., IO min., 20 min., 30 min., or 6o min.) it was pre-
sented with a second non-receptive female. The results are shown in Fig. io
as the mean number of head series in the second courtships of each group
of males. The line in this figure, at the level of 6.95, is the mean number of
head series obtained in a study of the male's first courtship (s = 2.09) on
the first day of its life and it represents the 'expected' behaviour of any such
group of males (BARRASS ig6ob). Comparison of the actual behaviour in the
second courtship with the expected behaviour in the first gives some idea of
the extent of the effects of this first courtship and the rate of 'recovery' from
these effects. The second courtship was most affected when it closely fol-
lowed the first. The rate of 'recovery' was rapid at first and then more
gradual.

Experiment Six.
Each male courted a receptive female and afterwards was observed with
a non-receptive female. The intervals between these observations were the
 301

same as in the previous experiment. The effect of courtship of the receptive

female can again be judged by comparing the male's behaviour towards a
non-receptive female in the second observation with the behaviour 'expected'
in a male's first courtship of a non-receptive female. The basis of this judge-
ment is the assumption that if the male's first courtship (receptive female)
had no effect, then the male would perform as many head series in its second
courtship (non-receptive female) as it would have performed in its first
courtship had the female been non-receptive. Similarly, if courtship of a
receptive female has an effect, then the extent of 'recovery' may be
considered greatest when the value obtained in the second courtship approx-
imates most closely to the value 'expected' in the male's first courtship.
Consideration of Fig. io suggests that these assumptions are usually correct.
The mean number of head series occurring in the second courtship (non-
receptive female) of the groups of 12 males used in this experiment are

Fig. io. The relation between the number of Fig. I The relation between the
head series in the male's second courtship number of head series in the male's
(non-receptive female) and the time which second courtship (non-receptive fema-
has elapsed since its first courtship (n o n - le) and the time which has elapsed
receptive female). The number of since its first courtship (r e c e p t -
males used at each time interval was 45 i v female). The dots show the
(V:, M), 10 (I M), 10 (3 M), 21 (10 M), mean behaviour of the 12 males used
35 (30 M) and io (6o M). The regression at each time interval. The regression
line was calculated from the original data line was calculated from the original
(b = 1.63) ; the crosses show the mean data (b = 1.20).
behaviour of each group of males.
302

SUCCESSIVE
OBSERVATIONS
Fig. 12. The number of head series in successive courtships of non-receptive females.
The interval between courtship was 30 seconds; except that following observation 20 it
was 3o seconds, 6o minutes or 24 hours. Means of 10 males in each case.
 303

shown in Fig. 11. The line at the level of 6.95 again represents the number
of head series 'expected' in a male's first courtship of a non-receptive female.
The second courtship was affected most when it occurred immediately after
the first. 'Recovery' was most rapid during the first few minutes following
courtship and was then more gradual.
The points shown in Figs io and 11are the means of the observations at
each time interval. The regression lines, however, were calculated from the
original data. The regression coefficients are b = 1.63 (Fig. io) and b =
1.20 (Fig. ii). Comparison of the slopes of the two regressions revealed
that the difference was not statistically significant (t = 1.5). Comparison
of the values of Y at a specified value of X ( X = 5.6 min.) also showed
that the difference was not statistically significant (t = 0.087). These
statistical methods and symbols are those given by SNEDECOR (1956).

Experiment Seven.
The behaviour of a male towards a succession of 20 non-receptive females
was observed (Part One). The interval between successive presentations of
the female was / min. After observation 20 a time interval was allowed
(Yí min. 6o min. or 24 hr) before the start of Part Two in which the male
was observed with successive non-receptive females as in Part One. Plan of
experiment:
A. i. 20 Non-receptive - Time - A. 2 20 Non-receptive
females Interval females
Ten males were used at each time interval and their behaviour is shown in
Fig. 12 as the mean number of head series occurring in successive observa-
tions. Part One of the experiment was performed in the same way with each
group of males so that comparable results are to be expected. Part Two was
also performed in the same way with each group of males and any dif-
ferences in the three cases must be attributed to the difference in time inter-
val between Parts One and Two (see table IV).
When the interval between observations 20 and 21 was % min. the male
courted few of the second 20 females and only performed a few head
movements and head series in these courtships, but there were a large number
of head movements in each head series. At the 60 min. interval, if the first
observation of Parts One and Two are compared, both the number of head
series (a.9) and the number of head movements per head series (5.12) in
observation 21 indicate considerable recovery. When, however, the whole of
Parts One and Two are compared, it is apparent that this partial recovery
is not continued throughout Part Two. This indicates that after one hour
the male was still considerably affected by its behaviour in Part One. When
304

the interval between Parts One and Two was 24 hours the male performed
more courtships, more head movements and more head series than it did in
Part One, but about the same number of head movements per head series:
recovery in this case was complete.

TABLE IV
The numbers represent means of 10 wcales, each of which was observed with
20 females in succession (Part i) and then 20 non-receptive females (Part 2).
A. i 2o non-receptive females - A. 2 2o non-receptive females
Time
Interval
B. I 2o receptive females -- B. 2 2o non-receptive females
The variable factors are i. the time interval between parts one and two and 2. the type
of female used in part one.
PART i.
Number of Number Number of head Nos, of head mvts
Courtships of head series in per head series
Movements observations in observations

PART 2.
interval in observations in observations

'
Experiment Eight.
This experiment differs from the previous one in that the first 20 females
were receptive, otherwise the plan of the experiment is the same. Plan of
experiment:
B. 1. 2o Receptive - Time - B. 2. 2o Non-receptive
females Interval females
The behaviour of each of the three groups of ten males is shown in Fig. 13
which refers to the mean number of head series occurring in successive
 305

observations and in table IV. Parts One are comparable since each group of
males was treated in the same way, Parts Two differ from each other only

OBSERVATIONS
SUCCESSIVE

Fig. 13. The number of head series in successive courtships of receptive females (Obser-
vations i to 20) and then non-receptive females (Observations 21 to 40). The interval
between courtships was 30 seconds; except that following observation 20 it was 30
seconds, 6o minutes or 24 hours. Means of 10 males in each case.
306

in the length of the time which elapsed since the end of Part One. The
number of courtships in Part Two cannot be compared directly with the
number in Part One since the stimulus-situations in the two cases were dif-
ferent. The previous experiment provides figures which can be taken as the
expected behaviour of a group of males towards non-receptive females in
Part One. In the present experiment the differences between this expected
behaviour and the observed behaviour in Part Two can be related to the
courtship of receptive females in Part One and to the duration of the interval
between Parts One and Two (table IV).
A male which has courted 20 receptive females shows a reduced tendency
to perform courtship movements when it is presented with 2o non-receptive
females immediately afterwards. After an hour has elapsed this tendency is
still apparent though it is not so marked and after 24 hours more courtship
movements are performed than would be expected of a male in the first part
of experiment seven (A. 1.).
Comparison of the effects of courtship of receptive and non-receptive
females upon the male's subsequent behaviour towards non-receptive females
is facilitated in table IV by the grouping together of the means relating to
each time interval. Considering any of the measures shown in table IV it is
readily apparent that courtship of 2o non-receptive females reduces a male's
response to further females more than does courtship of 20 receptive females.

DISCUSSION AND CONCLUSIONS

A waning in the responsiveness of the male M. vitripennis towards suc-
cessive non-receptive females has been described. The extent of this waning
was greatest when the time between courtships was small but even in this
case it was confined to the later parts of the behaviour. In all experiments,
when a female was present, the male chased it repeatedly.
In contrast, a male mated with many receptive females in a short time
(DnuTERT-WiLLrMZiK r93i; and the present work) and sometimes mated
twice or even three times in a single courtship. Mating in this insect does not
result in a reduced ability to repeat the same behaviour immediately.
When the behaviour of a male in its second courtship (non-receptive
female) was used as a measure of the effects of a first courtship (receptive
or non-receptive female), the results were not statistically different in the two
cases. Courtship of 20 females had different effects on the male's subsequent
courtship behaviour, however, according to whether or not these females
were receptive. It is concluded that some decrease in responsiveness does
result from a single courtship even when mating is not prevented but the only
indication of a waning in responsiveness to successive receptive females is the
 307

omission of the post-copulatory 'courtship' in later courtships. The waning

in a male's responsiveness to successive females presented at short time-
intervals was only marked when mating was prevented.
The waning in responsiveness towards non-receptive females can be
compared to the similar phenomenon observed in model (dummy animal)
experiments. The non-receptive female, like the model, precludes the per-
formance of the complete behaviour pattern but at the same time it lacks the
other imperfections of a model. The prevention of the later parts of a
behaviour pattern is an almost inevitable consequence of model experiments.
In view of the present results, therefore, studies on the waning of responsive-
ness in other animals should be treated with caution whenever models are
used. The waning in responsiveness to a model could just as easily be due to
the prevention of the later parts of the behaviour pattern as to the per-
formance of the first parts, and both factors are probably involved. Imper-
fections in the model might also be important but the present work provides
no evidence on this point.
The definitions of the terms 'consummatory act' and 'specific action
potential' (THORPE 1951) consider a waning in responsiveness to be due to
the performance of the consummatory act. In M. vitripennis a male's first
courtship does not end with mating (there is a post-copulatory courtship),
nor is the completion of the courtship associated with any marked waning of
responsiveness to further receptive females. The distinction between ap-
petitive behaviour and consummatory act is not possible therefore in this case
('I3ARRASS1960a).
BASTOCK, MORRIS & MOYNIHAN (1953) have suggested that the per-
formance of an innate behaviour pattern results in a particular series of
feed-back stimuli and that, at least in some cases, it is the receipt of this
correct feed-back that results in the satisfaction of an instinct. WATSON
(ig6i) has discussed the part played by reinforcement in 'drive-reduction'
and rejected the view that the reinforcing factor with respect to the primary
drives is the performance of a consummatory response. The absence of any
marked waning of responsiveness in the male M. vitripennis, when the feed-
back stimuli are appropriate, is also in contrast with this hypothesis.
Variations in an animal's response to a constant stimulus-situation are well
known in studies of innate behaviour (LORENZ I95o; ARMSTRONG I950; and
BASTOCK, MORRIS & MOYNIHAN 1953). A decrease in responsiveness has
been observed (TINBERGEN et al 1943; TINBERGEN & PERDECK 1950 and
HINDE I954a) when a stimulus-situation is presented several times in quick
succession, as has recovery during periods without release (HINDE 1954b).
These changes in responsiveness have been considered by previous authors
308

in terms of a using up of 'reaction specific energy' (LoRErrz 1950), a flowing

away of 'motivational impulses' (TINBERGEN 1950 and 1951), as a decrement
of 'specific action potentiality' (HINDE 1954b and c) and as 'habituation'
(THORPE 1959). ·
The changes in behaviour of the male M. vitripennis following courtship
cannot be explained by a using up of 'reaction specific energy' or 'motivatio-
nal impulses' since the waning is not marked unless mating is prevented.
HInDE (1959a and b) considers these terms to be obsolete, he rejects any
unitary concept of drive (see also WATSON 1961) and suggests the term
'specific action potential' as a direct measure of responsiveness to a given
stimulus. This 'specific action potential' depends on many variables :
"motivational" factors, learning processes, presence of compatible or in-
compatible response tendencies, etc. (HINDE 1959a). The concept of 'specific
action potential' is clearly intended to be descriptive rather than explanatory.
The present author has preferred to speak more simply of a decrease in the
duration of courtship, a decrease in the number of courtship movements or
a decrease in the frequency of courtship. The word 'waning' has been used to
indicate that all these things are happening together and the word 'recovery'
for the reverse changes.
THORPE (1959) has defined habituation so as to include the short-term
waning of a response. In M. vitripennis the waning of responsiveness was
never so complete that the male failed to respond (turning and chasing) to
the female. Also some decrease in responsiveness was noted even when the
complete courtship behaviour pattern had been performed. The changes in
the responsiveness of the male towards a constant stimulus-situation differs
in at least these two ways from the changes described as habituation in other
animals.
THORPE (1959) considered at least four types of response decrement to
be included under the general term "habituation". BAERENDS (1959)
reviewed ethological studies of insect behaviour and emphasised that various
mechanisms, on the way from sense organ to motor unit, can be responsible
for variations in the reaction to a constant stimulus-situation. LEHRMAN
(1956) also expressed this view.

ROEDER, Tozlnx & WEInNT (1960) demonstrated a widespread system

of spontaneously active efferent neurons in the male Mantis religiosa (L.),
controlled by inhibitory systems. They conclude that in insects inhibitory
control of local endogenous efferent activity by other nerve centres is the
rule rather than the exception. SHERRm·cTOrr (1933) and KENNEDY (1958)
have stressed the importance of inhibition as a nervous activity BAERENDS
 309

(ig5g) states that the importance of feed-back stimuli and of excitatory and
inhibitory stimulation of systems in the animal must be realised.
In M. vitripennis evidence for an endogenous central nervous influence
on the male's readiness to respond may be summarised:
a) BARRASS (ig6ob) noted characteristic variations in the number of head
movements in the successive head series of a courtship. Males were
grouped according to the number of head series in their first courtship.
In each group most head movements were performed in the first head
series and fewer in the second or third, there was then an increase fol-
lowed by a decrease in the last series prior to dismounting. Also, the
number of head series performed was found to be related to the number
of head movements in the first head series of a courtship. For example,
the courtships of males which performed most head movements in the
first head series included fewest head series and fewest head movements
in the last head series (and vice versa). In addition to these characteristic
variations, the mean number of head movements per head series was the
same in every group of males.
b) The waning in a male's responsiveness to successive non-receptive females
always involves a decrease in the duration of courtship and in the number
of head movements and head series but an increase in the number of
head movements in each head series. Recovery is the exactly opposite
changes.
In view of these regular variations, both within a single courtship and in
successive courtships, an endogenous central nervous influence on the male's
readiness to perform courtship movements is postulated.
A short-term response-specific effect has been demonstrated in the male's
courtship of a receptive female but the same behaviour may be repeated
immediately. Courtship of a non-receptive female has also been shown to
have an apparently similar response-specific effect on the male. A waning
in responsiveness to successive non-receptive females is attributed, however,
to an inhibitory stimulus-specific effect and in this case further courtship
is less likely. The male's courtship of successive receptive females indicates
that in this stimulus-situation the inhibitory effect is absent andlor mating
has an excitatory effect.
In interpreting these experiments on the courtship of M. vitripennis it is
not possible to think of a stimulus as allowing the discharge of a limited
amount of 'energy' from a functionally characterised 'centre' in the nervous
system. The experiments reveal an almost unlimited ability to perform
courtship movements providing the receptive female is always available. The
stimuli provided by a receptive female must direct nervous activity rather
310

than release 'reaction specific energy' or 'motivational impulses'. The waning

of responsiveness is largely the result of inhibitory effects which do not
operate unless mating is prevented.

SUMMARY
1. The method of dual quantification was used to study the effect of courtship of both
receptive and non-receptive females on the subsequent behaviour of the male Mormoniella
vitripennis.
2. The male's responsiveness to successive non-receptive females waned when the time
between presentations was short. The extent of this waning was less with longer time
intervals.
3. When many females were presented to a male one after another the male courted
almost all of them if they were receptive females but only a few if they were non-
receptive females.
4. A single courtship of either a receptive or a non-receptive female had a similar
effect on the male's subsequent behaviour and recovery occurred in a similar way.
5. Courtship of 20 non-receptive females reduced the male's response to further
females more than did courtship of 20 receptive females.
6. The significance of these observations is discussed with reference to the use of
dummy animals and to the recent ethological concepts of reaction specific energy,
motivational impulses, specific action potentiality and consummatory act.
7. An endogenous central nervous influence on the male's readiness to respond is
postulated. Courtship has a short-term response-specific effect (receptive or non-
receptive females) and an inhibitory stimulus-specific effect (non-receptive females).
With receptive females the inhibitory effect is absent and/or mating has an excitatory
effect. The stimuli provided by a receptive female must direct nervous activity rather
than release a limited amount of stored energy.

REFERENCES

ARMSTRONG, E. A. (1950). The nature and function of displacement activities. - Symp.

Soc. exp. Biol. No. 4, p. 361-384.
BAERENDS, G. P. (1959). Ethological studies of insect behaviour. - Ann. Rev. Entomol.
4, P. 207-234.
BARRASS, R. (1960a). The courtship behaviour of Mormoniella vitripennis Walk. (Hyme-
noptera, Pteromalidae). - Behaviour 15, p. 185-209.
- (1960b). The effect of age on the performance of an innate behaviour pattern in
Mormoniella vitripennis Walk. (Hymenoptera, Pteromalidae). - Behaviour 15, p.
210-218.
BASTOCK, M.; D. MORRIS& M. MOYNIHAN(1953). Some comments on conflict and
thwarting in animals. - Behaviour 6, p. 66-84.
COUSIN,G. (1933). Étude biologique d'un Chalcidien: Mormoniella vitripennis Walk. -
Bull. biol. 67, P. 371-400.
DAUTERT-WILLIMZIK, E. (1931). Einige Beobachtungen über das Geschlechtsleben der
Männchen der Schlupfwespe Nasonia brevicornis Ashm. - Zool. Anz. 93, p. 306-308.
HINDE,R. A. (1954a). Factors governing the changes in strength of a partially inborn
response, as shown by the mobbing behaviour of the chaffinch (Fringilla coelebs).
I. The nature of the response and an examination of its course. - Proc. roy. Soc.,
B. 142, p. 306-331.
 311

- (1954b). Factors governing the changes in strength of a partially inborn response,

as shown by the mobbing behaviour of the chaffinch (Fringilla coelebs). II. The
waning of the response. - Proc. roy. Soc., B. 142" p. 331-358.
- (1954c). Changes in responsiveness to a constant stimulus. - Brit. J. anim. Behav.
2, P. 41-55.
- (1959a). Some recent trends in ethology. - In KOCH(Ed.) Psychology: a study of a
science. Study 1, 2, p. 561-610.
- (1959b). Unitary drives. - Anim. Behav. 7, p. 130-141.
KENNEDY, J. S. (1958). The experimental analysis of aphid behaviour and its bearing
on current theories of instinct. - Proc. Xth Int. congr. entomol. 1956, 2, p. 397-404.
LEHRMAN, D. S. (1953). A critique of KONRAD LORENZ'S theory of instinctive behaviour.
- Quart. Rev. Biol. 28, P.335-363.
LORENZ,K. Z. (1950). The comparative method in studying innate behaviour patterns.
- Symp. Soc. exp. Biol. No. 4, p. 221-268.
- (1956). The objectivistic theory of instinct. In: L'instinct dans le comportement
des animaux et de l'homme. - Masson et Cie, Éditeurs, Paris. p. 51-76.
ROEDER,K. D.; L. TOZIAN& E. A. WEIANT (1960). Endogenous nerve activity and
behaviour in the mantis and cockroach. - J. ins. Physiol. 4, p. 45-62.
SHERRINGTON, C. S. (1933). The Brain and its Mechanism. - The Rede Lecture, Cam-
bridge. In: D. DENNY-BROWN Ed. (1940) Selected writings of Sir Charles Sherring-
ton, Epilogue p. 515.
SNEDECOR, G. W. (1956). Statistical Methods, 5th. Edn. - Iowa State College Press.
THORPE,W H. (1951). The definition of terms used in animal behaviour studies. - Bull
anim. Behav. 9, p. 34-40.
- (1956). Learning and instinct in animals. - Methuen, London.
- (1959). Learning. - Ibis 101, p. 337-353.
TINBERGEN, N. (1950). The hierarchical organisation of nervous mechanisms underlying
instinctive behaviour. - Symp. Soc. exp. Biol. No. 4, p. 305-312.
- (1951). The study of instinct. - 228 pp. Oxford, University Press.
-, B. J. MEEUSE,L. K. BOEREMA & W. W. VAROSIEAU (1943). Die Balz des Samt-
falters Eumenis (= Satyrus) semele (L.). - Z. Tierpsychol. 5, p. 182-226.
-, & A. C. PERDECK (1950). On the stimulus situation releasing the begging response
in the newly hatched herring gull chick (Larus argentus argentus Pont.). - Be-
haviour 3, p. 1-39.
WATSON,A. J. (1961). The place of reinforcement in the explanation of behaviour. Ch.
XI in THORPE,W. H. & O. L. ZANGWILL(Ed.) Current Problems in Animal
Behaviour. - Cambridge.
WHITING,P. W. (1958). Mormoniella and the nature of the gene: Mormoniella vitripen-
nis (Walker) (Hymenoptera, Pteromalidae). - Proc. Xth Int. congr. entomol., 1956,
2, p. 857-866.

RÉSUMÉ
I.La méthode de la quantification double fut employée pour étudier l'effet sur la
conduite postérieure du mâle de la Mormoniella vitripennis d'avoir courtisé des femelles
tant réceptives que non-réceptives.
2. La réaction du mâle à une série de femelles non-réceptives diminuait quand l'inter-
valle entre les confrontations était court. Cette diminution s'amoindrissait à mesure que
les intervalles se faisaient plus longs.
3. Quand de nombreuses femelles furent offertes au mâle l'une après l'autre, le
mâle les courtisait presque toutes si elles étaient réceptives; il n'en courtisait que quel-
ques-unes si elles étaient non-réceptives.