VISUAL RESPONSES IN LOCUSTS
George K. Wallace
A Thesis Submitted for the Degree of PhD
at the
University of St Andrews
1958
Full metadata for this item is available in
St Andrews Research Repository
at:
http://research-repository.st-andrews.ac.uk/
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10023/14870
A Theflif* P resen ted fo r the Degree
o f
Doctor o f Philosop^jy
o f
ST. ANDREWS UNIVERSITY
by
ProQuest Number: 10166621
All rights reserved INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a com plete manuscript and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
uest
ProQuest 10166621
Published by ProQuest LLO (2017). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States C ode Microform Edition © ProQuest LLO.
ProQuest LLO.
789 East Eisenhower Parkway P.Q. Box 1346
CERTIFICATE.
I c e r tif y th at George K. W allace, B #8c., has spent
the ecjuivalent o f nine terms a t research work under me, and th a t
he has f u l f i l l e d the con d itio n s o f Ordinance 16 (S t. jlndrews) so
th at he i s q u a lifie d to submit the acccog)a%yin,r th e s is in
a p p lica tio n fo r the degree o f Ph#D#
Natural H istory Department, Queen’s C ollege,
DUNDEE.
DECLARATION.
I hereby declare th a t the follow in g th e s ie im a record
o f req u ite obtained by me in the course o f research on V isu al
Responmem in Locuete and fu rth er th at the theeim i« ny own
c o n ^ s itio n and ha« not p reviou sly been presented fo r a higher
degree.
UNIVERSITY AND RESEARCH THAININCi.
I conmenced ay U n iv ersity mtudieg a t Queen’s C ollege (then
U niversity C o lleg e) Dundee, o f the U n iversity o f S t. Andrews in
1950. In 1953 I took a B .Sc. degree in Botary and Zoology and
in 1954 I was awarded F ir s t C lass Honours in zoology for research on th e ana tony o f the Sawfly cotnpound eye.
I was then awarded a Carnegie Scholarship and a Cross Trust
Scholarship to allow me to pursue research for the degree o f Ph.D.
( S t . Andrew’s ) , This research was carried out in the Hope Dept,
o f Entomology, Oxford, during the period 1954 - 1957.
I now hold a N u ffield *^enior Research Scholarship in the
CONTENT^..
Page.
ACKNOWl>:DGBH£IirrS
INTRODUCTION 1
MATERIAL 8
FORM PIJiCBPTIÜN
In trod uction 10
Experiments 12
The Importance o f Regions o f Contrast 13
Experiments w ith black o b jects on a wliite background 15
D iffer en t siz e d str ip e » 16
The Importance o f Contrast 16
The Importance of Area and Height 19
The Importance o f tlie V e r tic a l Edge 25
The Importance o f P o sitio n 26
The Importance o f a V e r tic a l str a ig h t edge as opposed 28
to an oblique str a ig h t edge.
The Im^x)rtance o f a str a ig h t v e r t ic a l edge 29
Tlie ^ p ossib ility o f an I.R.M . fo r a str a ig h t v e r t ic a l 31 edge
The Importance o f A r tic u la tio n 33
D iscu ssion 34
P3:ERIM?
Introduction 45
Method o f Measuring Peering Ang le s 46
Peering - a V isu al Response 47
Page
PEERING COKT’DPeering and D istance Estim ation 54
Peering - A G eneral Di<scu«^sicn 66
ANTENNA VvAVING
Introduction 70
Experiments 71
F lick in g 73
F lick in g - Dimcummion 75
B eating 76
Antenna Beating - D iscu ssion 86
The B io lo g ic a l ig n ific a n c e o f Antennal Movements aa 89 R eleased by v is u a l stim u li
Antennal movemnnts and d ista n ce estim ation 92
RKSPONSE TO ?jgOVEM£I^
In trod uction 93
F right Response to a moving ob ject 94
Optomotor Responses 97
Stim ulus 97
The E ffe c t o f a sm all part of the v isu a l f ie ld 98
Stim ulation o f d iffe r e n t parts o f the eye 100
Im or tance of the Type o f Movement IvO
D iscu ssion 101
Subject and O bject Movement 102
The von Buddenbrock Ejqperiment 103
The d iffe r e n tia tio n o f Subject and O bject movement 107
P rin c ip le 108
Experim ental con d ition s 109
Control 112
D iscu ssion 113
Page,
GENERAL DISCUSSION 116
T otal number o f ommatldia stim ulated and speed 119
o f image movement
A ttra ctiv en ess o f o b jects 120
Speed of image movement and p o sitio n on the eye 121
u b ject and O bject Movement 122
P o ssib le ela b o ra tio n of an in teg ra tin g system 125
d ealin g with v is u a l stim u li
CONCUJ^IOX 128
SUMMARY 129
APPf.NDIX 135
REFERENCES
ACKNOWL^;lX;Ki/HîiT^.
I to thank Dr, h'. P, Ihrarov, D irector o f the A n ti-iocu «t
Researcu Centre, for the opportunity to do th is work and P rof. G. C,
V ariey of the Entomology D ep t,, Oxford, fo r iiis kindness in afford in g
me f a c i l i t i e s in h is department. I am much Indebted to P rof, A. D,
Peacock on whose recommendation I obtained the Carnegie and Cross
Trust *^cholar*^tiipo which enabled me to pursue tid s research. For
th e ir generous fin a n c ia l a ssista n c e I wish to thank th ese foun dations,
My g r e a te s t debt of gra titu d e i s to Dr. F. w. Aaterhouse, queen’s
C ollege, Dundee, who, as ay su p erviso r, gave me much encouragement,
advice and h e lp fu l c r itic ism a t a l l stages o f tiie work.
I wish a lso o thank Dr. . î , E l li s fo r her encouragement and
advice.
Dr, Uvarov and Or, T, H. C, la y lo r of the A n ti-lo cu st Research
Centre gave me much encouragement in th is research . Through them I
was a lso given a c c e s s to much o f tiie lo cu st lite r a tu r e and in addition
awarded a grant for apparatus.
My thanks are a lso due to Mr, P, llur* er-tJones o f tiie Anti-Locu«t
Research Centre fo r ««upplyin^ lo c u sts in tlie d esired con d ition .
II^TRODUCTION.
In many in s e c ts the compound eyes are extrem ely prominent. They
are a pair o f rounded protuberances, the su rfaces o f which, under c lo s e r
exanriination are seen to be divided in to numerous hexagonal f a c t s . Each
o f th ese fa c ts i s the outer extrem ity o f a v is u a l receptor u n it - an
omma-tidium .
N aturally such conspicuous stru ctu res h/^ve attra cted much a tte n tio n
and Z o o lo g ists have been studying the functioning: o f th ese organs fo r
many y ea rs. Ever sin ce buuner’s c la s s ic tr e a tis e "Die P iiy sio lo g ie der
fa c e ttir te n Augen von ICrebsen und In sekten” vas published in 1891 there
has been an alm ost continuous output o f inform ation on the su b ject.
N ev ertlieless, although to-day th e lite r a tu r e i s le g io n , large lacunae
s t i l l remain.
Three main approacnes are u su ally made, the anatom ical, the p h
ysio-Iq g ic a l and the behavioural. The anatom ical work c o n sists in cu ttin g
se c tio n s o f the head and eyes and studying tiie d e ta ile d stru ctu re o f tlie
onsnatidia and th e ir arrange nent. The lite r a tu r e on tills asp ect o f the
su b ject i s too nuriorous to mention in f u l l but the work o f Nowikoff (1931)
and d e l P o r tillo (1936) are good examples. Very d e ta ile d stu d ies have
a lso been made of the anatosy of th e o p tic gan glia (C ajal and banchoz 1915; Zawarzin 19x4 ,in Reeder 1953).
In the pi^yaiological f i e l d , nerve im pulses have been recorded froxn
stim u li. This work has been r ecen tly reviewed very comprehensIvely by
VIrulff (1956).
The behaviour stu d ies c o n sist o f observing tlio beiiaviour o f Uie in ta c t aniifial u su a lly in response to con trolled experim ental con d ition s
o f v isu a l stim u latio n . In th is f ie ld tiio work o f d ertz, von F risch and
Buddenbrock are v e il known (se e bibliograpiiy fo r in d iv id u a l r e fe r e n c e s).
There are many other workers wtio w ill be referred to in tiie body o f the
te x t where th e ir work i s described in more d e ta il.
Two or more of tlie above asp ects are frequently coinbined, e .g .
Baumglàrtner (192 8), Bauers (1953) and a behaviour r e f le x may be studied in order to obtain p h y sio lo g ica l uieasurement , e .g . Hecht and %olf ( I927) measuring v is u a l acu ity and Schneider (1956) measuring sp ec tra l s e n s it
i v it y . In fa c t i t i s becoming in creasin gly clea r th a t, only with a
knowledge o f a l l the asp ects mentioned, can i t be hoped to ask meaningful
(questions with regard to the working of tiie con^und eye. Consequently
tiie r n su lts o f the anatom ist anci p h y sio lo g ist arf: b a sic to any behaviour
study such as tiie present one.
In tiie past the behaviour appro ch has e lic it e d much u se fu l in fo r
mation. From i t i t i s known th a t some in se c ts Lend to go towards lig h t
some aura from lig h t (Praenkel and C unn 1940), In other cases such as
bees and b u tt e r flie s , not only can the in se c ts perceive d ifferen ces in
lig h t in te n sity but they can d e tec t d ifferen ces between certa in forms
3.
perceive cer ta in colour* in some cases the s e n s it iv it y extending in to
the u lt r a - v io le t region o f the spectrum. Colour v is io n in In e c ts i s
reviewed by von Buddenbrock (1952)(se e a lso '^ulff 1956). Responses to
voK. Fr*i«<k p ola rised lig h t have be n demonstrated in bees {von F risch I9 4 9,j[l950)
in ants (Vowles 1950, 1954a, 1954b) and a saw fly (kfellincton 1953).
Mary varied responses have been obtained to movements in the environment
(se e referen ces in te x t p. 9 3 ) and some in s e c ts have been shown to e s
tim ate sh ort d istan ces very accu rately by a binocular method b a sic a lly
the same as th a t in Man (Balduo, 1926; F ried erich s, 1931). In sh ort,
such beiiaviour stu d ios have provided some id ea o f the v isu a l world of
the in s e c ts in qu estion . T his i s the aim o f the present in v e stig a tio n .
In th is presen t work the term " visu al response" ic defined as any
p iece o f behaviour pattern released or guided by v is u a l stim u li. This
th e s is i s an acccxint o f the study of a few such responses in the x^yn^ihs o f Uie D esert Locust, ( S ch istocerca ir e e a r ia (F orsk a l).^
Very l i t t l e i s known o f the part played by v isio n in the behaviour o f
lo c u sts in general and S ch istocerca in p a r ticu la r . I t i s b elieved
(Kennedy, 1939) th a t, in con d ition s o f moderate or low lig h t in te n s ity
(below 1,000 metre candles) and when not too a c tiv e , n/mpiis and adult^
2
o f S ch istocerca e x h ib it p o s itiv e p h ototaxis, i , e , they move towards a o 1. The in s e c t was o r ig in a lly described under G ryllus icrei-arius by z'orckal
in 1775. The name S ch istocerca fu*egaria was fix ed by Uvarov in I923
and the common name D esert Locust was coined by th a t author in 1928. H ereafter fo r convenience i t i s referred to as S ch istocerca or the D esert Locust.
4.
lig h t source (Fraonkel and Gunn 1940). Volkonsky (1939) showed th a t
S ch istocerca performs a non-locomotary o r ie n ta tio n to a lig h t beam aai
that th is o r ie n ta tio n i s accomplished p artly with the aid o f v is u a l
s tim u li, a movement o f the li,,h t source e lic it in g a coepensaiory move
ment o f the body. Again, the nynqphs of S ch istocerca are a ttra cted to
one another ( E llis 1956) and i t i s p o ssib le th a t v isica i plays a la rg e
part in tiiis as i t appears to do in the case o f the migratory lo c u st
Locuata ariLjaratorla m i^yatorioides ( E llis 1953 and personal connunication).
As can be seen from the above the inform ation a v a ilab le i s very fragmentary and a t b est g iv es l i t t l e idea o f the v is u a l world o f th is
in s e c t . There i s no dourt however th a t v isio n plays an important part in the behaviour of th is in s e c t as has been stron gly ind icated ty f ie ld
observations ( in p a rticu la r Kennedy, 1939; 1945).
As a su b ject for the study o f behaviour in r e la tio n to v is u a l stim u li
S ch istocerca i s id e a l. Since i t i s an exopterygote, as are the other
Orthoptera, there i s only s lig h t metamorphosis from qymph to a d u lt.
There are fiv e nymphal stages plus the adult sta g e. In the f i f t h nya^hal
s ta e , which is tiie one mainly used in tid s present work, the compound
eyes are w ell developed and p ossess about 7.685 f aoûts ( i . e . tiiis number
o f v isu a l receptors)(B ernard 1937). Roonwal (1947) showed th a t the eyes
were o f an anatom ical type (a p p o sitio n type) which i s tiiought to be par
tic u la r ly e f f ic ie n t fo r tlie perception o f forms and movements (%iggleeworth
5.
v isio n a n terio rly (th e se measure«nent8 are for the 5th in sta r nymphs,
W hittington, 1951 unpublished).
Apart from the advantages o f the w ell developed eyes the In se ct i s
la rg e , robust and easy to handle (len g th incroasos from 0 .33 to 0.45 mn
from the beginning to the end o f the 5th in s ta r . Uvarov 1928). This
large s iz e of the body and appendages makes i t r e la tiv e ly easy to observe
sm all d e ta ils o f behaviour and contributing to th is i s the fa c t th a t, when
undisturbed, the nymph Jtiovas f a ir ly slow ly. F in a lly , the f l i ^ t l e s s n e s s
in the nymphal sta g es makes the cons Lruction o f su ita b le experim ental
apparatus much e a sier than in the case of a fly in g in s e c t.
The tlie s is i s divided in to four experim ental se c tio n s as fo llo w
s:-I , Form P erception.
I I . P eering.
I I I . Antenna Vvaving.
IV. Responses to movement.
The form perception se c tio n stemmed d ir e c tly from the r e s u lts o f
previous workers. Even i f tiiere were no ^ e a t development o f form d is crim ination in S ch istocerca to the ex ten t th a t there i s in bnes, never
th e le s s some response to o b jects was extrem ely lik e ly . The suppcx't fo r
th is c(xnes from the work of Hundert nark (1937b) who showed tiia t even le p
i-dopteran c a te r p illa r s with only groups o f o c e l li have some degree o f farm
d iscrim in ation . Sawfly c a te r p illa r s p ossessin g two la rge o c e l li show an
a ttra c tio n to v e r t ic a l black/w hite edges (' a lla c e 1954 unpublished). I t
did not seem unreasonable, th erefo re, to expect th at some response to
6.
form would be presen t in S ch istocerca where the Go»^)ound eyes are so w ell developed.
The "peering" and "antenna waving" sectio n s stemned from the form
perception se ctio n in th at both peering and antenna w vin g were p ieces
o f behaviour observed during the o rien ta tio n and a ttra c tio n to forms.
Both responses were shocn to be v is u a l responses and were studied in d e ta il
Tiiroughout the work th ere was an increasing eu^g e stio n o f the impor
tance o f movement as a stim ulus and th is culminated in a b r ie f gen eral
survey o f t h is in the la s t se c tio n . I t w^s in f a c t shown th a t a l l the
responses observed could probably be explained on tlie b asis o f a response
to novement. The e f f e c t s o f d iffe r e n t so r ts of movement were stu d ied .
A behaviour appro ch of ttiis kind i s fraught w ith one major i n i t i a l d if f ic u lt y . T his i s that tlie perceptual world of the in se c t stu d ied i s
probably very d iffe r e n t ftrom tiia t o f the observer. Nor, in the behaviour
o f Man, v is io n plays such an impijrtant r o le th at i t i s d if f ic u lt to
approach any o b je ctiv e study of in s e c t v isio n wiUiout a t le a s t a few pre conceived id e a s. To guard a g a in st t h is the a p r io ri assumptions (and a
few are in e v ita b le ) must be kept to a minimum. In ttd s th e s is t h is has been done witii grea t strin gen cy. For example, i t i s f i r s t shown th a t
the in se c ts are a ttra cted to region s o f ccantrast before any study i s made
o f responses to complex fo m s. ^gain, the o b jects and patterns used
throughout th is work have been tw o-dim ensional. Imost nothing i s known
about the part played by th ree-d im en sion ality in the v is u a l environment
case o f human v is io n , on knowledge o f p ersp ectiv e, on in te rp re ta tio n
o f shading and on d iffe re n c es o f focu s. o in :e there i s , as y e t, no
evidence o f the operation of tljese mechanisms in in se c ts the perception o f th ree-d im en sion ality in t h is case cannot be assumed. F in a lly , black
and grey o b jects liave been used again st v^iito backgrounds for th ere i s
no good evidence a t present to siiow th a t the in s e c t can p erceive colour.
The th e s is tlm s r eso lv e s i t s e l f in to the quostion of what the
in s e c t can do with a U n ite d number o f sim ple v isu a l stim u li, the use
to which v is u a l inform ation i s put and the methods o f obtaining t h is .
O ften the r e s u lt i s su rp risin g and the responses a; parently q u ite complex.
Im p lic it in t^iis approach i s the concept of the compound eye as a navig
atin g instrum ent fo r , in the l a s t an alyses, the fu n ction o f the sense
organs i s to enable the animal to move about fr e e ly in i t s envircximent. The r e s u lts give soins id ea o f the v isu a l environmmit o f S ch istocerca , rej^-aria although time has not perm itted the study o f colour or p olarised
lig h t , two stim u li which may v e i l form an important part o f th is environ ment. As w ill be pointed o u t, certa in e t a i l s of th is v is u a l envircmment
which i s th a t o f a crawling in s e c t (in the nymphal stages used) may d iffe r
iiïço rta n tly from the environment of a fly in g in s e c t. I t is probable, however, th a t many of the p r in c ip le s in v estig a ted here apply to a l l
in s e c ts w ith w e ll developed compound eyes and some may extend over even
8.
MATERIAL.
Sch istocerca r remaria i s a polyinorpiiic sp ec ies e x istin g in a range
o f forms which d i far m orphologically and b io lo g icfiU y , Tho extremes
o f th is range have been c a lled phases. In one phase the in s e c t i s a
s o lita r y grasshopper, th is i s th e phase s o lit a r ia . In the other i t i s
a ty p ic a l swarming lo c u s t, th is i s th e phase , r e ;a r ia . (Uvarov 1928).
By d e fin itio n a lo c u s t i s a grasshopper with a swarming phase (Uvarov 1928).
In the phase iXeFSria th e ny iphs arm dark in colou r, in the la te r in sta r s
the general c o lo ra tio n i s yellow or orange-yellow w ith a strongly pro
nounced Mack p attern (a fte r Uvarov 1928). The ey s are darkly pigmented.
Nymphs o f the phase s o lit a r ia are pale green in colour with sometimes
tra ces of black markings. The eyes are pale with v e r tic a l s tr ip e s .
For a d escrip tio n o f th e eyes o f th e two phases se e Roonwal 19-4^7* In
the present in v e stig a tio n only numphs of the phase Fregaria were used. The In sects were obtained from the A nti-Locust Research Centre,
London, u su ally e ith e r as f i r s t in sta r nymphs or as eggs la id by Fre^ a ria
a d u lts. (In a few cases 4th or 5th in sta r n;noq)h8 were obtain ed ). The
nymphs were reared in crowded c o n d itio n s, about 6X) f i r s t in sta r hymphs
to a cage 18 inches square by 12 inches tiigh. By Uie 5th in sta r tlxere
were approximately 300 in se c ts per cage.
Two opp osite w a lls o f the cages were o f wire • ause to allow a ir to
c ir c u la te fr e e ly . The other w alls were o f hard-board and the top was o f
g la s s . The cages were kept o v ein ig h t in a constant temperature room
9-the constant tenporature room sin ce space was lim ited and o9-ther experi
ments were in progress. However, ty means o f e le c t r ic lig h t bulbs sus
pended above the cages the temperature was maintained at a reasonably
high le v e l (26 - 32<^).
The in s e c ts were fed each evening with fresh grass standing in water.
Under the con d ition s described the in s e c ts developed tlie ty p ic a l gregaria co lo r a tio n .*
Care was taken to ensure, as far as p o ssib le , th a t the nymphs were
not over a c tiv e or h igh ly e x c ite d when they were used fo r experim ents.
This was achieved by ju d iciou s feed in g . I t was found tliat i f nymphs
were te ste d in the morning (10 a.m. to about 12 noon) they did not require
any feedin g other ttian the rou tin e feedin g tlie previous evening. Nymphs
te ste d la te r than th is were fed two hours before te s tin g . D espite tiieae
precauti(xi8 v a r ia tio n s in e x c it a b ilit y were s t i l l encountered probably
due p artly to the respO'“8e o f d iffe r e n t in d ivid u als to handling or to
d iffe re n c es in physiology o f the in d iv id u a ls.
The experim ents were performed in the constant temperature room a t
289c, Whenever p o ssib le the only l i . ht present was t l» one used to
illu m in a te the apparatus. At tim es when, unavoidably/, other experim ents
ere in progress any apparatus described here was screened o f f from the
Influence of extraneous lig h t . There was no indie? tio n th at such ex ter
n al lig h t in terfered w ith the responses o f the in s e c ts .
10.
FORM PERCEPTION.
Introduction,
One o f the main featu res o f the v isu a l f ie ld i s the presence o f
o b je cts. In se v er a l cases i t has been shown th a t in s e c ts respond to
o b jects in the environment, and th a t tiie responses are mediated p a rtly
by v is io n o f the form o f th ese o b je c ts, e .g . bees and b u tte r flie s are attra cted to flow ers (von F risch 1914; U s e 1932), and stic k in s e c ts
are a ttra cted to black str ip e s rep resen ting stems (Kalimis 1937) . In the sim p lest a n a ly sis, o b jects appear merely as areas o f the
v is u a l f i e l d , the r e fle c te d lig h t from wtiich d iffe r s cp ian titatlvely and/or q u a lita tiv e ly from the lig h t r e fle c te d from tiia ir surroundings.
The presence of o b je cts in a v is u a l f ie ld can thus be detected by a
sim ple photo-receptor s e n s itiv e enough to record sm all changes in lig h t
in te n s ity , but the perception o f the form o f such o b jects req uires a
more coo^lex recep tor. Indeed, Buddenbrock (1935) se e form perception
a b ilit y as th e sharpest d is tin c tio n between higher and lower compound
e y e s.* Thus he s ta te s th a t, in some b e e tle s and other in se c ts with
even le s s w ell-developed e y e s, no reaction to form could be obtained,
nor was tiiere any other evidence o f form perception in th ese anim als.
To show th a t an animal can p erceive form i t i s not s u f fic ie n t to show th a t i t can see o b je c ts, i t must be shown th a t the animal can d is tin
gu ish one from another.
• i l .
Amongst in s e c ts t h is a b ilit y to discrim inate between forms has,
however, been c le a r ly demonstrated in bees (von F risch 1914; Baumgibrtner
1928; Herts 1929a, 1929b, 1931, 1933, 1934c, 1935, 1937), and in b u tter
f l i e s ( U s e 1932) , The a b ilit y o f the wasp P liilanthus to d istin g u ish between arrangements o f sm all o b jects (van Beusekom 1948) i s a lso evidence
o f form d iscrim in a tion .
Of form perception in Orthoptera l i t t l e i s known. ’ Kalnus (1937)
showed th a t s tic k in s e c ts ( D ixioous aorosus) were a ttracted to v e r tic a l
b la c k /w ilte boundaries and to black str ip e s on a w hite background (se e
a lso Hundertmerk 1937c). W illiam s (1954) worked w ith the grasshc^pers
Ciiortippus p a r a U e llu e (Z e tte r ste d t) arxi Comphoeeriopus rufus ( 1 .) and
concluded th a t they possessed th e a b ility to d iscrim in ate between forms,
but h is evidence i s not d eta ile d and ^ives l i t t l e In d ication as to
wiiich p rop erties o f the fig u r es are involved .
I t i s g en era lly accepted th a t tlie e ffic ie n c y o f a con^pound eye fo r
forming the image o f fig u r es i s measured to a g rea t ex ten t by the number
o f tiie fa c e ts . Thus, . . ." th e r e t in a l im age.. . . c o n sists o f a mosaic of
poin ts o f l i g h t . a mosaic which w ill be coarse or fin e depending on
the number of fa c e ts per u n it area". (% igglesworth 1953). The
"apposi-*
tion" type o f compound eye i s thought to g iv e a sharper image than the
su p erp osition type of eye. The former i s the type o f eye found in the
Uymphs and ad u lts o f S cliistocerca gregaria (Roonwal, 1947) and i t has
numerous fa c e ts ( e .g . 4 t in sta r nyn^h 6,460; 5th in sta r nymph 7,685
12.
p o ssess the r e q u isite apparatus fo r form d iscrim in ation (a t le a s t a t the le v e l of the r e tin a ). The aim o f the present work, th e re fo re , was to
in v e stig a te form perception a b ility in th e nymphal stages o f S ciiisto cerca
and to conqjare the r e s u lts w ith th ose wtiich had fircviously been found fo r
bees where the b a sis o f form p e r c ^ tio n had been studied in g r e a te st
d e ta il ( Hertz 1929» e t c . , Wolf and Zerrahn-Wolf 1936), I t was a lso
considered a matter of in te r e s t whether or not the b a sis o f form percep
tio n and ttie a ttr a c tiv e p rop erties o f forms were the same fo r a alking
in s e c t (lo c u s t nymph) as fo r a fly in g in s e c t (b e e ).
The problem was thus to fin d ,^ f i r s t l y , i f lo c u st uyn^hs were
a ttracted to d b jects in th e environment and seco d ly , i f the in s e c ts could d istin g u ish between the forms of th ese o b je cts.
Experiments.
Apparatus and Procedure.
The b asic method used in a l l the form perception experim ents was to
place an animal in the cen tre o f a large c ir c u la r arena (diam eter 2 f t . ) whose v e r tic a l 10" w alls were decorated with the p attern s to be te ste d .
The subsequent behaviour of th e animal was observed. In t h is way the
animals were allowed to choose spontaneously between the forms presented.
L ighting was from an overhead source so arranged as t o ^ive eqaal
illu m in a tio n over a l l parts o f the arena with no heating e f f e c t . A ll
13.
experiments wore perfor^ned in a constant temperature room a t 2ô9C, The animals were fed before te s tin g so as to reduce th e ir a c tiv ity *
Further d e ta ils o f the nymphs used in each p a rticu la r t e s t , i . e . in sta r
and number o f tim es te s te d , an- given under the separate experim ents.
The Importance of Regions o f C ontrast.
Experiment 1: Prelim inary experim ent. Equal areas o f black and w tdte.
S ch istocerca gregaria i s knovn to be p h o top ositive (F raenkel, 1929,
Kennedy 1945), as are many other eer e adidae (See Grasse 1923, Chapman
1954, 1955) . I t was thus p o ssib le th a t, i f the nymphs were placed in a heterogeneous environment c o n sistin g of equal areas of black and w idte, they might show an a ttr a c tio n to the w hite patches. On the other hand
E llis (1953) shoved tlia t in the case of lo c u s t m igratoria, when not s it t in g with oth er nymphs, the in s e c ts preferred to s i t near black
ob jects rather than white ones. This was on an almost wiiite background.
I t was p o ssib le th erefore ttiat in tlifp case the S cliistocerca nymphs might
be found to choose tlie black areas o f the environment. Again, i t was
p o ssib le tJiat they would bo a ttra cted to black/w hite boundaries, c . f .
s tic k in s e c ts (Kalous 1937). The present experiment was desi«_ned to
t e s t these p o s s ib ilit ie s .
Forms. The circum ference of the arena was 72" and tliis len gth of
w a ll was divided in to e ig lit eq asl se c tio n s, four black and four wiriito.
Each se c tio n thus measured 9" broad by 10" high and a black altern ated with a w liite. Thus, to an anim 1 in the centre o f the arena, the
Ik.
The animals had, tiie re fo re , an equal chance of approaching a black or a
w hite area. The flo o r was of p la in wtdte paper.
Animal*». These were $ t h in s ta r nym phs of the normal greg aria
phase.
Behaviour. An animal placed in the centre o f the arena remained
station ary for a short period (u su a lly - J m inute). I t then swayed
the f l'ont part o f the body «low ly from sid e to « ice sev era l tim es before
sta r tin g to w ait. This swaying o f the boyd i s c a lle d "pperin^,"
(Kennedy 1945), end i s more f u lly described la t e r (p .4 5 ). The
ani a ls were found to approach the w a ll o f the arena, in some cases
stopping from time to tin .e, peering and o cca sio n a lly changing d ir e c tio n . The above de«cribes tlie ty p ic a l beh-viour seen in a l l ttie form
perception experim ents,
Reading*. %hen the in s e c ts nad rfcached the w a ll, the p oin t a t
w ic h they f i r s t touched i t was noted. The readings were divided in to
three catégorie® , (1) animal® going to the black areas; (2) those going
to w lilte areas; ( j) tho«e going to tlie v e r tic a l black/w hite boundaries
between the area®.
The experiment was repeated «eversl tim es. r u n s were per farmed
with }0 d iffe r e n t anim els.
R e su lts. The r e s u lts are : iven in ta b le i . They show c le a r ly th a t
the M ajority o f the insect® were attracted to the v e r t ic a l black/w hite bou»
15.
insec'ts W0r<5 a ttra cted ©itJioi* to the black or the v h lte , eqgial nuBibera going to each.
Conclusion, I t vas concluded th a t lo c u st nymphs are attracted to
v e r t ic a l b laek /vtiite boundaries,
E^ineriments v ith black objecta on a white background.
The f i r s t ejqperiment had sh ow th a t the in se c ts sore attracted to
the region s of con trast in the v isu a l f ie ld . They had, hovever, been
presented e ith equal areas o f black and s h ite , a situ a tio n rather far
removed from the normal. The natural v isu a l environment c o n sists o f
r e la t iv e ly sm all o b jects seen aga in st a r e la tiv e ly larger background.
I t vas th erefore decided to decrease the s iz e o f the black areas and in
crease the s iz e o f the white so as to present the animals w ith sev era l black o b jects on a large w hite background.
From now on, a l l the form perception ejqperiments described, follow ed
e x a ctly the same procedure. The background was supplied by the white
w all of the arena on to which the o b jects were fix e d so as to touch the
flo o r . The ob jects were cut out o f th in card, painted black with
in oian ink and fastened to the w all w ith concealed s tr ip s o f se llo ta p e .
The method o f measuring the r e la tiv e a ttra c tiv e n e ss o f the forms
was to present tiie animals with e ig h t o b je c ts, four o f one shape and
four o f anottier and to count the number of animals going to each. The
o b jects were placed a t equal in te rv a ls and an ob ject o f the one shape
T
T
= Pe « ring
P = Tall object
S = Shr/t object
F ig . 1 .
16
.
was rotated with resp ec t to the paper flo o r ond a t in te r v a ls the paper
flo o r was renewed. In t h is way the p o s s ib ilit y o f th e animals follow in g
scen t t r a ils of aqy d escrip tio n was removed.
Experiment 2i D iffer en t sized s tr ip e s .
Forms, Four t a l i black s tr ip e s 10” high X 2” broad and four sh o rt
black str ip e s 5** high X 2” broad.
Animals. 5th in sta r nymphs o f the normal pregarla phase. 40 runs
were made with 40 d iffe r e n t anim als.
R e su lts. The numbers going to the t a l l and short str ip e s were
noted, as w ell as the numbers going to the v e r tic a l edges o f th e s tr ip e s .
An animal was counted as going to an edge i f i t h it the w all w ithin .5 ”
of an edge. The r e s u lts are presented in ta b le 2,
The r e s u lts show th a t the in s e c ts were a ttra cted to the black fig u r e s ,
showing a h igh ly si^?nificant preference for the t a lle r fig u r es (p ^ .0 1 ),
Furthermore, the in s e c ts were c le a r ly a ttracted to the v e r tic a l edges o f
the s tr ip e s . F ig , I shows a specimen o f the tracks recorded.
C onclusions. I t was concluded th a t lo c u st nymphs are attra cted
to t a l l black recta n g les on a w tlto background. The t a lle r the rectan g le ive
the (lore a ttr a c te d i t i s .
The Importance o f C ontrast. Experiment 3» Black s tr ip e s and rrey s tr ip e s .
The aim o f th is experiment was to t e s t further th e Inqportance o f
17.
given th e choice o f black s tr ip e s and grey s tr ip e s o f the same s iz e .
3a, Forms. Black s tr ip e s 10” t a l l & 2” broad. Grey s tr ip e s (N 5,5
o f the M unsell s e r ie s ) 10” t a l l X 2” broad. This grey i s f a ir ly dark. Animals. 4 th in sta r* phase lure* a r ia . 10 animals te ste d each 5
tim es,
R esu lts. The r e s u lts are given in ta b le 3 and show ttjat the grey
used in th is experim ent was as a ttr a c tiv e as the black (,5 ^ p ^ .3 ) . 3b, In tills experiment a lig h te r grey was used.
Forms. Four black s tr ip e s 10” t a l l X 2” broad. Four grey str ip e s
(IK ID 7 /1 o f the îiu n se ll s e r ie s ) 10" t a l l X 2” broad,
Anl iials. 5th in s ta r , phase rrep.aria. 10 animals each te s te d 5
tim es.
The eiQ)eriiQent was repeated with 4th in sta r nymphs, 20 were te ste d
each 5 tim es.
C ontrol. In the co n tro l experiment the in s e c ts were presented with
8 lig h t grey s tr ip e s .
R esu lts. The r e s u lts are given in ta b le 4, In con trast to the
r e s u lts obtained in experiment 3a, the r e s u lts now obtained showed th at
there was a h igh ly s ig n ific a n t preference for the black s tr ip e s ( p < ,0 1 In
each c a se ). I t was not the case th a t the grey s tr ip e s were not v is ib le
* There did not appear to be any important d ifferen ce between the be
haviour of the 4th and 5th sta g e rymphs. Further observation
con-firmed t h is . U su ally, however, 4th stage qymphs were never used ex
UNIVERSITY OF ST. ANDREWS
D J / j m .
From
The Dean of the Faculty of Science, David Jack, M.A., Ph.D., F.R.S.E.
COLLEGE GATE, ST. ANDREWS.
5th March, 1958.
Dr# J #M# Dodd,
Department of N atural H istory,
ST.SALVATOR'S COLLBOE.
Dear Dr. Dodd,
The following Committee has been
appointed to examine the Ph.D. Thesis submitted
by Mr. G.K. W
allacet-Dr. Waterhouse (Convener),
professor Callan,
Dr. Dodd.
As Dr. Waterhouse has already seen the
Thesis, I should be obliged i f you would return
the enclosed copy to Mr. Ritchie a fte r you have
seen i t .
Yours sincerely .
18.
to the in se c ts fo r the con trols shoved th at. In the absence of the black s tr ip e s , the animals orientated p e r fe ctly v e il to the lig h t grey ernes.
Conclusion. Experiment 3 demonstrated th a t, on a v h ite background,
the darkness (blackness) o f an ob ject i s an important property af actin g
the a ttra c tiv e n e ss o f the o b ject. However, at th is point a cautionary
word must be in serted . I t i s known from the work of Hertz (1937) that
su rfa ces, which to our eyes appear grey, appear to bees as d e fin ite
colou rs. I f the same i s tru e of lo cu sts then i t i s p o ssib le that the
greys used in the present experiments appeared to the in se c ts as d is tin c t colou rs. On the other hand, the f i r s t grey widch vas used was equally
as a ttr a c tiv e as the black and i t was a daik grey wiiich suggested th at
the in se c ts were indeed responding to the con trast between o b ject and b ackground. furtherm ore, observation c le a r ly showed th at the in se c ts were o rien tatin g to the v e r tic a l edges o f the o b jects, i . e . the regions
o f con tra st, in the case of both the black and grey str ip e s,
V»heth€3T or not lo cu st nymphs see greys as colours cannot be decided without a fuztber se r ie s of d e ta ile d experim ents. Although no readings were taken o f i t in tliis experim ent, observation showed th a t the animals
were again orien tatin g to the v e r tic a l edges o f tiie str ip e s rather than to th eir cen tres. The fin d in gs thus suggested th at the black ob jects
were a ttr a c tiv e because, when seen on the white background, they had a
19.
The Importance o f Area and H eight.
The r e s u lts o f the wq)eriment with t a l l and sh ort str ip e s (ex p t, 2)
had now to be reccxisidered. In the li^ h t of the obvious importance o f
the v e r tic a l ed^^ i t was tem pting to suppose th a t tije greater a ttr a c tiv e
n ess of the t a l l str ip e s was due to their p ossessin g longer v e r t ic a l
edges. However, although i t had been shown tlia t, #ien approaching à
s tr ip e , the in s e c ts o rien tated towards the ve t i c a l edge, there was no
evidence to show th a t the edge of the str ip e was what had i n i t i a l l y
attra cted the in s e c ts . The t a lle r str ip e s d iffe re d from the sh orter
ones in two ways, f i r s t , they had a larger area (20 aq .” as a g a in st 10 sq .')
and secondly they had longer v e r t ic a l ed^es. Botii str ip e s were black
and th erefore the sharpness of th e ir ed^es was the same. Their breadtlxs
were equal. Thus, e ith e r the greater area o f the t a l l s t r ip e s , or
th e ir longer v e r t ic a l edges could be the cause o f th e ir greater a ttr a c tiv e
n ess.
I t was tlierefoi'e decided to t e s t sep arately the importance o f th e
two p ro p erties area and h eig h t. This was to be one hy presen ting the. in s e c ts f i r s t w ith s tr ip e s o f constant h eight but d if f é r a it area and
second w ith s tr ip e s o f constant area and d iffe r e n t h eigh t.
Experiment 4. S tr ip e s of eq u al heiidit and d iffe r e n t area.
A .a. Forms. Black s tr ip e s 10** t a l l X 2” broad. Black str ip e s 10” t a l l X
4” broad. In th e ta b le o f r e s u lts the former s tr ip e s are c a lle d "narrow**
20.
tw ice th a t o f the narrow,
Anima ls , 5th in s ta r , phase ; reg a ria . 10 animals te ste d each 10
tim es. The eiq>eriment was repeated with 4th in sta r nynphs of th e phase
jgregaria. 10 animals were te s te d each 5 tim es.
R esu lts. The r e s u lts are given in ta b le s 5& and 5b. T h ^ are
extre-nely su rp risin g , fo r , contrary/ to showing e ith e r a choice o f the
la rger o b jects or an equal ch oice of both, the animals stiowed a higtxly
s ig n ific a n t preference for the narrower o b jects (p ^ ,01 in each c a se ).
C onclusion. I t was c le a r from the r e s u lts that the o b jects o f
sm aller area were c e r ta in ly the most a ttr a c tiv e . The creation now was,
what made the narrower s tr ip e s so a ttr a c tiv e by comparison? In both
experiments the track s of the animals had been recorded on p ilo t sh eets
and i t was q u ite clea r tiiat ^lere again the a n in als were o rien ta tin g to
the v e r t ic a l edges o f tiie s tr ip e s . The only d iffe re n c e between tt»
v e r t ic a l ed^es of the narrow s tr ip e and those o f the broad str ip e was
that in th e former ti.e two edges were c lo ser together w hile in the la tt e r
they were further apart. I t was thus p o ssib le th a t herein la y the
explanation of the greater a ttra c tiv e n e ss of the narrower s tr ip e . The
argument could be put as fo llo w s
:-I t has been suggested th a t wlien an animal o rien ta tes to an o b ject
i t turns towards tiie o b ject and brir^ s the image o f i t on to a fix a tio n
region o f the eye - the region o f the forwardly d irected ommatidia.
I f , due to any d ev ia tio n in the in s e c t 's tra ck , the image moves th is
21.
th e image once more cm to the fix a tio n région . (Fraenkol and Gunn 1940). I t was p o ssib le th a t the animals preferred to o r ie n ta te to the narrow
s tr ip e because i t allowed botli v e r t ic a l edges to be "fixed" by tlje anterior
part of the e y e, whereas in a broad str ip e the v e r t ic a l edges were too
w idely separated to be so " fixed " ,*
4 .b , Subdivision o f the broad s t r ip e s .
I f , as suggested above, the in s e c ts were choosing the o b ject whose
v e r tic a l edges we e c lo s w togeth er and not j u s t the o b ject o f sm aller
area, i t should be p o ssib le to in crease the a ttr a c tiv e n e ss of the broad
str ip e s by adding a v e r t ic a l black/w h ite mdgs between the two o u tsid e
edges. The presen t experiment was to t e s t t ills .
Forms. These were the same as the forms used in the previous experi
ment, except th a t each broad s tr ip e was subdivided by a very narrow wnite
str ip e (.25" broad) running v e r t ic a lly down th e cen tre, i . e . black
s tr ip e s 10" t a l l X 2" broad; black str ip e s 10" t a l l X 4" broad + narrow cen tr a l w liite s tr ip e .
Animals. 4th in s ta r , phase ; rsj>aria. 20 anim als, each te s te d 5
tim es.
C ontrol. The c o n tro l was done by rer,oving the narrow w hite s tr ip e s
and thus returning to th e o r ig in a l situ aticxi o f broad and narrow rstripes.
R esu lts. These are t,iven in ta b le s 6a and 6b,
The co n tro ls showed th a t there was a preference fo r th e narrower
str ip e s as has been found p rev io u sly , l^hen th e buad str ip e was sub-As w ill be seen la t e r , there i s a sim pler explanation fo r the a ttr a c t
22.
d ivid ed , th e b ia s was a ltered and there was eq^al choice o f the broad
and narrow s tr ip e s , ( . 2 ^ p ^ .1 ) . The d ifferen ce between th is and th e
co n tro l experixaent vas liigh ly s ig n ific a n t (p < .0 1 ),
C onclusions, In th e situ a tio n s presented to the in s e c ts in th ese experim ents 4a and 4b, i t had thus been shown th a t la rg er area did not
make a fig u re a ttr a c tiv e i f , a t the same tim e, the d ista n ces between
the v e r tic a l edges was too g r ea t. The presence a n d ^ s itio n o f th ese
edges appeared to be more important than area. I t vas s t i l l p o ssib le , o f course, th a t, in sm aller fig u r e s whose breadths were not so great as
to separate the v e r tic a l edges too w id ely, area might prove to be im
p ortan t, The fo llow in g experiment te ste d t h is .
Experiment 5. S tr ip e s of medium breadth and very narrow s tr ip e s .
Forms. Black s tr ip e s 5" t a l l X 2" broad (broad s tr ip e s ). Black
str ip e s 5" t a l l X ,25" broad (narrow s tr ip e s ) .
The broad thus p ossessed four tim es the area o f the narrow ones,
however, the broad s rip ns were only 2" broad and i t had already been
shown th at str ip e s o f th is breadth were not too broad and were attractive
to the in s e c ts , (expt^, 2 , 3 , 4 ),
Animals. 4th in s ta r , phase i reg a ria , 20 anim als, eacn te s te d 5
t i e s .
R e su lts. These are presented in ta b le 7 . The animals were found
to show a preference fo r ttie broader str ip e s (,0 2 ^ p ^ ,0 1 ),
C onclusions. Th« broader s tr ip e s , in th is experim ent, now proved
23.
forward above, to exp lain the Importance of the width apart o f the
not
v e r t ic s l eddies couldj^apply in tiiis ca se, fo r here the in s e c ts were
choosing the wider s tr ip e . I t seems reasonable to suppose th a t, in
tills ca se, th e animalr made th e ir choice in favour o f the la r g e s t area.
P o ssib ly , a t a dishnce, the broader str ip e i s more d is tin c t than the
narrower one, which w ill only subtend a very sm all h orizton a l an gle. In au ming up, then, i t can be sta ted th a t, when a l l s tr ip e s are
o f equal h e ig h t, a str ip e of interm ediate breadth (2") i s preferred
e ith e r to a very broad str ip e (ex p t, 4a) or to a very narrow s tr ip e
(ex p t, 5 ), the important fa cto r in the f i r s t case being th e d istan ce
apart o f the v e r t ic a l edges and in the second case th e area.
The orij in a l purpose o f the experiments on breadth had been to t e s t
the e ff e c t of increased sr* a fo r constant height in an attempt to ex
p la in the a ttra c tiv o n ss o f a t a l l str ip e 10" X 2" over a sh ort str ip e
5" X 2" (ejqpt, 2 , p ,6 ). I t was now seen th t the e ff e c t o f area was
unpredictable and th at area d id n t seem to be as important as the
v e r tic a l s r ip e edges. The r e s u lts o f experiments 4& and 4b had again
pointed to th e inqportance o f th ese edges (p.2.0-11 ) , The importance
o f the h eight o f t h is v e r tic a l ec4_e was now stu d ied ,
Experiment 6 . S trip es o f equal area and d iffe r e n t riei^'ht.
Forms. T a ll black str ip e s 6" t a l l X 1" broad. Area « 6 sq ," .
Short black str ip e s 3" t a l l X 2" broad. Area — 6 sq".
Animals. 5th in s ta r , phase r ez a ria . 20 anim als, each te ste d 5
24.
C ontrol. A co n trol experim ent was done by removing the t a l l
s tr ip e s and presen ting the in s e c ts with the four sh ort ones.
R esu lts. Those are given i ta b le 8, The animals showed a very
s ig n ific a n t preference for the t a lle r s tr ip e s . The sh orter s tr ip e s were
alm ost e n tir e ly ignored (on ly 9 h it s ) although the con trols showed th a t,
in the ab-ence o f the t a lle r s tr ip e s , the sh orter s tr ip e s were a ttr a c tiv e
to th e in s e c ts , (p C .0 1 ).
In the experim ental runs th e behaviour o f the in se c ts made i t c lea r
th a t, in many c a se s, they were making th e ir choice from the centre o f the
arena, turning so as to face tlie t a lle r s tr ip e s . In ono case a qymph
was seen to face a p a rticu la r t a l l s tr ip e , the a ena was then rotated
s lig h t ly (th e flo o r remaining sta tio n a r y ). A fter approximately 2 seconds
tlie animal reo rien ta ted in the new d irectio n o f the t a l l s t r ip e . In the
con trols the in s e c ts were again seen to face the str ip e s from the cen tre,
showing th a t the sh ort str ip e s were p e r fe c tly v is ib le from the centre
o f the arena where the choice was made.
Co n clu sion . I t i s concluded tiia t, under the Cvnditions reported,
the fig u re o f larger v e r tic a l edge i s the more a ttr a c tiv e .
Experiment 7 . O bjects o f equal area and equal perim eter.
I t was ju s t p o ssib le th a t, in the previous experim ent, the animals
chose the t a lle r str ip e because i t possessed a larger périmétaijthan the
sh ort str ip e (14" as again st 10") although th is was thought u n lik e ly ,
Experiment 7 was designed to c la r ify th is p o in t.
25.
recta n g le 2" X 1" w ith longer edges h o rizo n ta l. Of n e c e ssity the
o b jects wer*^ s n a il fo r the la r g e st sid e could not be more than 2” , other
w ise th e breadth of the h orizon tal fig u r e would con p lica te m atters. (In
la te r fig u r es (e^qpts, 11, 12, 13) the breadth was 3" but th is was n ot ioM
portant sin ce a l l the fig u r es were o f the same breadth).
Animals, 20 5th in sta r nymphs each tested 5 tim es,
• R e su lts. The r e s u lts are given in tab le 9. They chow th a t, fo r
o b jects of equal area and perim eter, th e more a ttr a c tiv e are those with
the la r g e st v e r tic a l edge. The number o f m isses was la rg e , undoubtedly
due to tlie sm all s iz e o f the fig u r e s , fo r , from the cen tre o f the arena,
the angle subtended by the o b jects was very sm all.
Importance o f the v e r tic a l edue.
A ll the €Otperiments performed had demonstrated, by process o f elim
in a tio n , the unique importance of the v e r tic a l edges o f the o b jects used.
The experim ents iiSd been in s tig a te d in an atten p t to exp lain why t a l l
str ip e s should be more a ttr a c tiv e than short s tr ip e s , as found in the
i n i t i a l experiment (ex p t. 2 ) , There was now no doubt th at tbe t a l l s tr ip e s were more a ttr a c tiv e , not by v ir tu e of th e ir larger area (sin c e
a large area i s not always a ttr a c tiv e , exp t, 4, and th e t a lle r str ip e i s
chosen even when the areas o f the s tr ip e s are the same, expt 6 ) , nor
v ir tu e of th e ir larger perim eter (sin c e a d ifferen ce in perim eter i s n ot
n ecessary, exp t, ?) * but by v ir tu e o f th e ir longer v e r t ic a l ed^es ( a l l
26.
4b p .2.1 ) ,
Henceforth a tten tio n was focused on the q u a litie s o f ttd s edge.
These were (a ) p o sitio n , (b) v e r t ic a lit y , and (c ) str a ig h tn e ss.
The importance o f p o sitio n .
The term p o sitio n should be further c la r ifie d . Perhaps th is can
b e st be achieved w ith referen ce f i r s t to human v is io n . In kan, when
the body and head are in the e r e c t posture the oyes look along a lin e o f
sig h t p a r a lle l to th e ground, i . e . h o rizo n ta l, ÜVhen the person i s
standing on a f l a t h o rizo n ta l surface with the eyes a t f i r s t looking
h o r iz o n ta lly , any o b ject so situ a te d that the head or eyes have t o be
t i l t e d up to lo<^ a t i t must be situ a ted v e r tic a lly w ith refer*ence to
the plane on which the observer i s standing. Now, because o f the
stru ctu re o f the compound ey e, in s e c ts do not have to move th e ir heads
to see such an o b je ct, i t w ill be seen hy cm natidia more towards the top o f the eye. %ae i t p o ssib le then, th a t an ob ject wtiose image appeared
on the upper ommatidia, was more a ttr a c tiv e tiian a sim ila r o b ject seen
by the more h o rizo n ta lly d irected ommatidia? In otiier words, did an Imsge appearing on the more upwardly d irected ommatidia si^_nify a t a l l
o b ject for the in sec ts? (TMiat mtght be c a lle d an IHU r elea sin g a ttr a c tio n
to a t a l l o b je c t, Tinbergen 1951 ) . The p o s s ib ilit y was te ste d in the
follow in g experiment.
Experiment 8. O bjects in high and low p o sitio n s.
Forms. Black r ecta n g les 3" t a l l X 2” broad, bases on th e flo o r
o f the arena. 3 la c k r e c ta n g les 3" t a l l X 2" broad, bases 3" above -s'’ / "
27.
the flo o r r e sp e c tiv e ly as had been the case with th e o b jects 6” X 1" and
3" X 2" in exp t, 6 where the t a l le r o b jects had been p referred ).
Animals. 5th in s ta r , phase j reg a rla . 10 animals each te ste d 5 tim es.
R e su lts. The r e s u lts ( t a l e 10) showr^d th at a s ig n ific a n tly greater
number o f the in s e c ts went to the lower o b jects (,05>p^02) •
C onclusion. The r e s u lts o f th is e^qperiinent are d i f f i c u lt to in te r p r e t
The fig u r es show th a t a m ajority o f the in s e c ts went to the lower o b je c ts.
However, i t was o cca sio n a lly seen th a t in s e c ts lleading for one o f the
higher o b jects turned asid e Wien c lo se to i t and went to the low ob ject nearby. I t i s not }X>seible to say ex a ctly how o ften th is happened, i . e .
how o ften the primary orientatlcm was to the higher o b je c t, fo r o ften one
cannot t e l l w ith any c e r ta in ty to which o b ject the in s e c t i s o rien ta tin g
u n til i t i s f a ir ly c lo se to i t ,
PYoTi the behaviour o f the In sect in t h is experiment a reasonable
conclu sion would seem th a t probably both o b jects were eq ually a ttr a c tiv e
from a d ista n ce* , and th a t the apparent preference for the lower o b ject
was due to the in s e c ts * lo sin g th e ir way* ?ticn c lo se to the higher o b ject
t
and r eo r ie n ta tin g to the lower one, Y ith referen ce to the o r ig in a l
question o f the a ttr a c tiv e n e ss o f a t a l l ob ject i t would th erefore appear
appear th a t th is o b ject was a ttr a c tiv e , not because i t s edge stim ulated
more dor s a lly d irected ommatidia but because i t possessed a longer
v e r tic a l edge, Whether the lo s s of o r ie n ta tio n when c lo se to the higher
o b ject i s due to the o b je c t’s foresh ortening in th a t p o sitio n , or to tlie
a: ■ i... ; v : ^ ^
' r mëâM
’■ "^':V . 'f
4
4
4 b
■ I
F ig, 2a,
F ig , 2a and 2b:diagram summarising the r e s u lts of ^ ^ form perception experim ents w ith lo cu st nymphs, ' "^he numbers r e fe r to th e numbers of th e experim ents in th e te x t and th e + in d icates th e p re fe rre d fig u re of each p a ir ,
No^c ftu 4 " "4"'" “*-*
bwv.. 9 . "m, "
b « c a o ir 0^ îKc 0|- tke (>Kotôy*|>k.
