Skip to main content

Full text of "Laboratory equipment for psychological experiments"

See other formats

ti !tit:hi:t! : 

IS I 111 




University of California 









Professor of Psychology and Director of the Psychological 
Laboratory at Yale University 












The description of exercises given in Volume II of this 
series was of set purpose dissociated from the description 
of apparatus, in view of the fact that the same laboratory 
exercise may be tried with a very great variety of dif- 
ferent kinds of material aids.. These material aids can 
be mastered by the student very much better in the pres- 
ence of the apparatus than through any written description. 
After he has mastered a given piece of apparatus he should 
be called upon to give a description of the construction 
and working of the apparatus in his report. The present 
volume is, accordingly, not designed primarily for the 
student. If the student is called upon to prepare his own 
apparatus as well as work out the exercises, this book 
will be of advantage to him; it may very properly be used 
therefore, by graduate students who are preparing to teach 
the subject. Its chief function, however, will be in offer- 
ing suggestions to those who wish to give demonstrations 
or to teach laboratory courses. 

No large expenditure of funds is necessary in order to 
secure a sufficient equipment with which to conduct a 
course in experimental psychology. The apparatus neces- 
sary for this course can with a few exceptions be constructed 
with the aid of carpenter's tools. Full lists of apparatus 
are given on pages 243 to 249 The Yale Psychological 
Laboratory is prepared to supply all of the necessary 




pieces of apparatus for this course to any one who may 
wish to purchase them. Correspondence is invited from 
any who wish to make purchases, and a detailed price list 
will be mailed on application. If modifications of designs 
furnished are desired, these modifications will be intro- 
duced wherever it is possible, at the direction of the pur- 
chaser. In general it is so highly desirable that the equip- 
ment of small laboratories for demonstration and practical 
laboratory courses be promoted, that one of the important 
functions of the Yale Laboratory in connection with its 
graduate work is the provision of material for the work 
of its students and others of like interest who become 
teachers of the subject. 

No effort has been made in the following pages to de- 
scribe all of the different pieces of apparatus available 
for the various lines of experimentation; only those are 
described which are judged in the author's experience to 
be serviceable for the purposes here considered. Fur- 
thermore, unless apparatus is distinctly the work of a 
single individual no effort has been made to give an histori- 
cal account of the way in which it has been designed 
and modified by successive workers. A practical man- 
ual of laboratory equipment is all that is aimed at. 
Many of the figures are copied directly from the catalogues 
of makers and are acknowledged in the text where they 
appear. The other obligations of the author to investi- 
gators are numerous; many are acknowledged in the text, 
many require no special acknowledgment because the 
apparatus and method have been adopted in common use. 

It is a pleasure to make special acknowledgment of 
the contributions of two gentlemen who make it their 
special business to design and construct psychological 
apparatus. The workshop of the Yale Psychological 
Laboratory had for a number of years the very competent 


services of Mr. Charles Herbert Smith. Mr. Smith con- 
structed many of the pieces described in this book and drew 
in many cases the figures. His successor, Mr. Teeuwen, 
prepared others of the drawings. 

With the appearance of this volume the series of three 
text-books originally planned is completed. Subsequent 
volumes on the application of psychology to education 
and cognate subjects are in preparation and will be an- 
nounced more fully later. 

C. H. J. 

New Haven, September, 1907. 






A Apparatus and Procedure . . 16 

B Results 22 

C Supplementary Experiments 24 


A Apparatus and Procedure 31 

B Results 41 

C Supplementary Experiments 42 


A Apparatus and Procedure . . 47 

B Results 56 

C^-Supplementary Experiments 58 


A and B Apparatus, Procedure, and Results 64 

C Supplementary Experiments 68 


A and B Apparatus, Procedure, and Results 78 

C Supplementary Experiments 87 


A Apparatus and Procedure 98 

B Results. 103 

C Supplementary Experiments 104 


A Apparatus and Procedure 106 

B Results 115 

C Supplementary Experiments 116 


A Apparatus and Procedure 118 

C Supplementary Experiments 119 





A Apparatus and Procedure 121 

B Results 124 

C Supplementary Experiments 125 


A Apparatus and Procedure 128 

C Supplementary Experiments 132 

A Apparatus and Procedure 137 


A Apparatus and Procedure 160 

B Results 164 

C Supplementary Experiments 164 



A Apparatus and Procedure 169 

B Results 171 

C Supplementary Experiments 171 


A Apparatus and Procedure 175 

B Results 180 

C Supplementary Experiments 181 


A Apparatus and Procedure 182 

B Results 196 

C Supplementary Experiments 196 


A Apparatus and Procedure 201 

B Results 203 

C Supplementary Experiments 203 


A and B Apparatus, Procedure, and Results 206 

C Supplementary Experiments 211 



XVIII. EFFECTS OF PRACTICE (a) Impression Factors 213 

C Supplementary Experiments 214 

XIX. EFFECTS OF PRACTICE (6) Motor and Perceptual 

A Apparatus and Procedure 217 

B Results 217 

C Supplementary Experiments 219 


A Apparatus and Procedure 220 

B Results 220 

C Supplementary Experiments 220 


A Apparatus and Procedure 224 

B Results 227 

C Supplementary Experiments 228 


A Apparatus and Procedure 229 


A Apparatus and Procedure 231 

B Results 237 

C Supplementary Experiments 238 


A Apparatus and Procedure 239 

B Results 240 

C Supplementary Experiments 240 

ESSES 242 



INDEX.. . 253 





A PSYCHOLOGICAL laboratory has two closely related 
functions. It should be supplied, in the first place, with 
apparatus that will make possible the demonstration 
of certain typical forms of experimental investiga- 
tion before the elementary class which is pursuing an 
introductory course. For example, if one is discussing 
binocular vision as a form of perception it is very ad- 
vantageous that the members of the class should have 
the opportunity of experiencing by direct observation 
some of the binocular combinations which can be made 
with an ordinary hand stereoscope. If the class is large, 
the apparatus should be duplicated so that each member 
of the class may have a reasonable opportunity to make 
the observation for himself. On pages 250-252 is given a 
list of the various demonstrations which can be used 
with the author's Psychology, General Introduction. .Such 
demonstrations as these pave the way for more elaborate 
experiments on the part of the student. 

In so far as the demonstrations cover the same ground 
as the laboratory exercises, which constitute the second 
function of a psychological laboratory, the equipment 
for demonstration will be the same as the equipment 
for the various individual exercises. There is, however, 
a certain equipment which is especially needed for purposes 


of demonstration and is not required for the laboratory 
exercises to be discussed in detail later. First among 
these are either charts or facilities for lantern projections. 
Charts have the advantage of keeping before the student 
for a long time the outlines which it is intended to im- 
press upon him. The best form of cheap chart is a solar 
print, such as is made by Sprague and Hathaway, 36 
Bromfield Street, Boston, Mass. These photographers 
copy directly any figure or negative on charts, which 
they supply for fifty cents each. The advantage of a 
lantern, on the other hand, is that it is possible for 
any one, with a little practice, to copy from any book 
the figures which may be desired for demonstration 
purposes. After meeting the initial expense of the lan- 
tern, the further acquisition of material is relatively cheap 
as compared with the purchase or preparation of charts. 
If possible an electric arc lantern should be secured. 
A laboratory of modest equipment can dispense with any 
of the more elaborate lanterns and will find a small hand 
lantern serviceable for many practical purposes other 
than those of demonstration. Somewhat better than 
charts or lantern slides for the demonstration of parts 
of the central nervous system and the organs of sense, are 
models. These give the student in plastic form a very 
much better idea of the shape and relation of the parts 
than do flat drawings. The most desirable models are 
suggested in the list on pages 250-252. 

Special demonstration equipment may also include cer- 
tain large pieces which are designed for use with large 
classes. These special pieces need not be discussed here. 
For most laboratories the same piece will serve for both 
demonstration and individual experimentation. 

Turning from the first function of the psychological 
laboratory as a source of demonstrations for the elementary 


course, the second and more highly specialized function 
is to provide the material for individual laboratory exer- 
cises to be carried out by the students themselves. On 
pages 243-249 there is given a list of the apparatus with 
which the experimental course described in the Labo- 
ratory Manual can be conducted. On the left side of the 
pages are mentioned those pieces which are required, 
on the right those which are desirable. This list can be 
indefinitely enlarged as the resources of the laboratory 
permit. On the other hand, it can also be reduced. In 
case the left-hand list is reduced, it will be found that the 
completeness of the experiment suffers. Any one who 
can not try the complete experiment should not hesitate 
to do whatever he can. The results may be very sug- 
gestive, even if they lack something of the fulness and 
accuracy called for in the text. 

There are certain general types of equipment which are 
not discussed in connection with the special exercises. 
First, some equipment for drawing and, secondly, some 
equipment for shop work are very desirable. A great deal 
can be done by the instructor or by the student in the way 
of preparing simple apparatus for the laboratory ex- 
ercises. Thus, it will be shown in the description of the 
equipment necessary for the first exercise that all that is 
absolutely essential is a small number of cards, a measur- 
ing rod, a ruling pen, and a simple board tray. The ability 
to draw simple line figures is easily acquired, and should 
be cultivated in order to prepare such figures as those 
under discussion, and also in order to produce the graphic 
representations of results which are to be discussed later. 

It is highly desirable that there be in the psychological 
laboratory provision for various electrical connections. 
The greater the variety of connections available, the 
more useful will this equipment be. If possible, connec- 


tions should be secured with a direct current, such as is 
used for incandescent electric lights. In case such a 
current is not available, substitutes may be sought in 
various forms of battery currents. If batteries must be 
used, the Edison-Lalande batteries are to be recom- 
mended as the most permanent and productive in current. 
If the incandescent lighting current is used, it will be 
found that this current is supplied at a potential which is 

FlQ. 1 

From Studies. From the " Yale Psychological Laboratory," Vol. IV 

too high for use in ordinary apparatus. A convenient 
method of reducing the current is a lamp battery described 
by Professor Scripture and represented in Fig. 1. In 
this battery the current is drawn through the wires E 
and F. Following the circuit shown in the diagram at 
the right, we find that from F the current passes first 
through the lamp at A. The lamp at A may be of 
any desired candle-power. It acts as a resistance and 


reduces the current supplied through E and F to any de- 
sired quantity. Thus, if a lamp of 32 candle-power is 
employed for A, the current will be reduced to about 1 
ampere; a 64 candle-power lamp gives about 2 amperes, 
and a 100 candle-power lamp gives about 4 amperes. 
After passing through the lamp A the current may be 
carried along the path 1, 2, and may either be turned 
through B or be carried on to C. If carried to C and there 
drawn off through wires connected with the apparatus, the 
current supplied to the apparatus will be, after it is well 
established, 1 ampere. The current will then pass out 
through the series of connections marked 1, 1, 1, E. 
This simple connection has certain marked disadvan- 
tages, especially at the moment the current is turned off, 
when there is likely to be a spark which may burn out any 
delicate connections. In order to take up the spark, also 
in order to reduce the potential of the current throughout 
the experiment, it is better to leave open to the current 
the second path B, 2, G. 

At B a small lamp is inserted. This lamp should have 
an amperage equal to that of the large lamp. For example, 
if the large lamp A is 32 candle-power, the small lamp 
should have an amperage of 1 ampere. The potential 
of the small lamp may be relatively small, ranging from 
6 to 12 volts. If now with the connection B, 2, G made, 
the current is drawn as before to the apparatus from the 
plug at C, the circuit B, 2, G, constitutes a shunt, and any 
spark which is generated at the moment the appa- 
ratus is disconnected is taken up by the shunt circuit. 
The shunt circuit will also act so as to reduce the poten- 
tial of the current used throughout the apparatus. Further- 
more, the relation between the two circuits passing through 
the plugs C and G is such that when a connection is made 
through an apparatus of low resistance at C, the current 


at G is virtually broken, and, conversely, when C is broken 
G is made. This relation may be used by connecting 
apparatus at both C and G; or a simple connecting plug 
may be inserted at G, in which case the one source of sup- 
ply will be through C. The following table, prepared by 
Dr. Scripture, is in use at the Yale Laboratory: 


Lamps used in the batteries 

on Lamp 

Trade Name 


110 volts 100 c. p. 4 



110 volts 100 c. p. 3J 



110 volts 64 c. p. 


110 volts 32 c. p. 


110 volts 16 c. p. 


110 volts 8 c. p. 



on Lamp 

Trade Name 


8 volts 4 amperes 


8 volts 4 amperes 

8 volts 4 amperes 


12 volts 3 amperes 


12 volts 2 amperes 


12 volts 1 ampere 


12 volts 0.7 ampere 


10 volts 1 ampere 


6 volts 1 ampere 


20 volts 16 c. p. 

Results of various combinations of lamps. 

Lamps used. . . . 

Am An Ao Bm Bn 



Cm Cn 

Potential in volts 

.9 5 7 7 5 



4 3 

Max. cur. in amp.. 

.4.0 4.0 4.0 3.5 3.5 



1.9 1.9 

Lamps used .... 

.Co Cp Cq Cv Dp 



Dt Du 

Potential in volts 

.4 7 10 25 4 



10 5 

Max. cur. in amp. 

1.9 1.9 1.9 1.9 1.0 



1.0 1.0 

Lamps used .... 

.Dv Eq Er Es Et 



Fs Fv 

Potential in volts. 

. 15 3 5 7 4 



4 6 

Max. cur. in amp. 

.1.0 0.5 0.5 0.5 0.5 



0.3 0.3 


The lamp battery is relatively uneconomical in its use 
of the current as compared with a transformer through 
which the current from the main line may be reduced be- 
fore it is used in laboratory apparatus. But a transformer 
is an elaborate piece of apparatus, and it permits only 
a relatively small range of modifications of the current at 
different points in the laboratory. The lamp battery is 
portable and very much cheaper in initial cost than the 
transformer. It is capable of readjustments by changing 
the lamps and yet is sufficiently fixed in its output so that 
the student will not derive from it a current large enough 

FIG. 2 

at any time to endanger apparatus by any mistakes in 
connections which he may make. For all these reasons 
it will be found to be a very convenient adjunct to labo- 
ratory equipment. 

The connections with this battery and with the current 
can be made by means of the ordinary screw sockets or by 
means of contacts of the following type. The plug for this 
contact is represented in Fig. 2, and consists of a hard rub- 
ber block bored with two holes. Through these two holes 
pass the wires D and E, which are to carry the current. 
At one end of the hard rubber block these holes are en- 



larged so as to make place for two screws A and B, which 
are set down into the hard rubber. These screws clamp 
down the ends of the wires. If the free ends of the screws 
are split and the two parts of each of the ends are spread 
slightly apart, they will spring firmly into the socket, which 
is represented in section in Fig. 3. The socket is just 
the converse of the plug. It is made of hard rubber and 
is bored with holes for two wires. These holes are en- 
larged so as to receive two metallic cups which are sup- 
plied with screws at their base and with shallow slots at 
their upper end, these slots being for a screw-driver. The 
cups are used to clamp down the supply wires, and screw 

FIG. 3 

far enough into the base so as to be below the sur- 
face of the block. They are thus protected, so that it is 
impossible for anything to come in contact with them ex- 
cept a pair of metallic points which can be fitted into the 
two cups. The hard rubber plug carrying these cups is 
supplied at its edges with two extensions by means of 
which it can be screwed into the wall. The advantage 
of such a plug and socket as this is that the current can 
be instantly made and broken without any screwing or 
unscrewing such as is common in the ordinary lamp plugs. 
A fourth general type of equipment which will be found 
very convenient in the laboratory consists of table clamps. 


S-clamps, rods, and holders. A form of universal table 
clamp which is very convenient is represented in Fig. 4. 
It consists of a screw clamp to be fastened to the table and 

a split ball which is designed to 
carry the rod. This ball can, 
when undamped, be so adjusted 
as to allow the rod to stand 
at any desired angle. When 
clamped firmly in position by 
means of the screw lever shown 

rio. 4 

in the figure, it holds the rod at 

any position in which it has been set. A second very con- 
venient table clamp is represented in Fig. 5. The two jaws 
of this clamp can be set at any desired 
distance from each other along the rod so 
that they can be fitted without difficulty 
to any table or block. Such a clamp 
holds a rod only in a position perpen- 
dicular to the table to which it is 
clamped. The S-clamp or right-angle 
clamp is represented in Fig. 6. By 
means of this, rods of various sizes 
can be fastened at right angles to each FIQ 

other. Rods can be secured either from 
the manufacturers or they can be bought in stock from any 
hardware dealer and can be cut into desired lengths by 
means of a hack-saw, which should be 
one of the tools in the workshop of the 
laboratory. An equipment of clamps 
and rods can be utilized for a great 
variety of purposes. For example, the 
Wheatstone stereoscope described on page 82 can easily 
be set up with rods and clamps, no special construction 
such as that described on page 82 being required. 


A general question of procedure which may be discussed 
at this point, rather than in connection with any of the 
particular experiments, is the question of the observer's 
general attitude toward the experiment. It will be found 
with all beginners that there is much distraction in the 
conditions under which the experiment is made. Further- 
more, there is a natural feeling on the part of most students 
that a psychological experiment is a test of their individual 
ability and that they must, in order to do themselves 
credit, carry out the test with the greatest degree of rapidity 
possible and with the highest grade of attainment in num- 
ber of striking judgments. To a certain extent this atti- 
tude must be fostered, for the success of all psychological 
experiments depends upon the highest possible degree of 
attention which the observer can give to the problem in 
hand. On the other hand, precautions may very properly 
be taken in order to avoid excessive anxiety on the part 
of the observer, lest he should fa'il to comply with the de- 
mands of competition in the. experiment. It is better for 
an observer to recognize that regularity of reaction or of 
judgment indicates a higher type of efficiency than a few 
very rapid and accurate judgments in the midst of a 
general series which is uneven in its average. Above all 
things the observer should free himself from the dis- 
position to adhere to any preformed theory of what the 
results of the experiment should be. Some experimenters 
have believed that the best way to avoid preconceptions 
on the part of their observers is to keep them wholly igno- 
rant of the subject of the experiment and also of their own 
results. There can be no question that under certain 
circumstances it is highly desirable that the observer 
should not be confused by a knowledge of the results. For 
example, in most practice series it is better that the ob- 
server should give himself wholly to the task of practice 


rather than that he should have his attention divided 
between practice and thought about how rapidly or how 
slowly he is progressing. In other cases, however, it is 
better to give the observer definite information, at least as 
to the general direction in which the experiment is tending. 

For example, in the experiments on sensation inten- 
sities, especially in the determination of initial thresholds 
(see page 128), it is very commonly advantageous to tell 
the observer the direction in which variations are being 
undertaken. If this is not done, the condition should, so 
far as possible, be arranged, as indicated under the de- 
scription of the method of right and wrong cases, so as 
to take advantage of the chances which result when the 
observer makes a blind guess. The observer's attitude, 
it will be seen from this discussion, is always a factor in a 
psychological experiment. 

No rigid rule can be laid down for the treatment of 
the observer. He will improve during his contact with 
psychological problems whatever procedure is followed; 
and he will ultimately find that a wholly unbiased opinion 
is the most productive for the investigation. But these 
matters of attitude can be cultivated only with experience, 
they can not be forced upon the observer at the outset. 
The more experience an observer can have with regard 
to a problem and its methods before he begins the in- 
vestigation, the more rapidly he will become a mature 
observer. Procedure with knowledge, as it is technically 
called, is in general more advantageous than procedure 
without, except as indicated above, where the problem 
under consideration is one of practice and does not in- 
volve any knowledge on the part of the observer of the 
degree or direction of his change through practice. 

In this connection it may be well to point out that the 
results obtained by beginners are very frequently so in- 


coherent and different from the general results reported 
in the papers of trained experimenters that the instructor 
must constantly supplement the student's results by ad- 
ditional information or references for reading, in order 
to make the work productive. If the student obtains re- 
sults which will justify only a very broad general conclusion, 
there should be no effort to force upon these vague results 
any greater refinement than they readily justify. It 
should be explained to the student that the results which 
he finds reported in the more elaborate investigations are 
the outcome not only of the method which he is pursuing 
but of a higher degree of training in observation and re- 
action. He should be encouraged to recognize indi- 
vidual differences as important matters in all psychologi- 
cal investigations. He should be encouraged in con- 
nection with each experiment to give as full an account 
as possible of his own subjective condition. This will 
in many cases throw light upon the character of the re- 
sults which he obtains. Further discussion of the matter 
of exactness of psychological results, especially in so far 
as quantitative methods are in question, has been under- 
taken in the general introduction of the Laboratory Manual. 
When the student has obtained his results, the next 
matter is the presentation of these results. In general 
it will be found advantageous to bring together all the 
results of an investigation whenever possible in a graphic 
representation. A graphic representation has the ad- 
vantage over a table that it presents at a single glance 
the whole series of results in their relations. The student 
should be made familiar with the use of coordinate 
paper and with the principles of reduction to various 
scales. For example, the following table was secured by 
an investigator who carried out a practice series of measure- 
ments with the Miiller-Lyer illusion until the illusion was 



much weakened. This table was brought together in the 
curve represented in Fig. 7. 

FIG. 7 

Figure 10 cm. in length, oblique 3 cm., at angle of 45. 


Avg. 111. 
in m.m. 



Avg. 111. 
in m.m. 


April 26 



May 27 
June 1 




May 2 






















































The table contains the material which is expressed in 
the curve, but it is obvious that the comprehension of the 
whole matter is made very much simpler by the graphic 
representation. Students should be encouraged to pre- 
pare compact tables and then devise various methods of 



presenting these tables in curves. Attention should be 
called to the fact that the size of the figure is a matter of 
total indifference, provided the units of measurements in 
the horizontal are all of exactly the same kind and that 
the units of measurement in the vertical are in turn homo- 
geneous. It is sometimes advantageous to emphasize one 
characteristic or the other of a given curve. In that case 
small units may be used for the horizontals and larger 
units for the verticals. Thus, the same table as that pre- 
sented in Fig. 7 is condensed in its horizontal dimensions in 

. 8 

Fig. 8. In the author's Psychology, General Introdujction, 
graphic representations of this type will be found on pages 
340, 372, and 373. 

The most convenient unit of space measurement for 
all laboratory exercises is the centimeter and its multi- 
ples and subdivisions. The student should be made ac- 
quainted as soon as possible with these units, and meter 
rods should be provided in the general equipment of the 
laboratory. It is sometimes desirable to provide the stu- 
dent with measures of the metric system for his use 
outside of the laboratory. Millimeter coordinate paper 


may be purchased and may be cut into small strips, if it 
is desired to attain this end at very small expense. This 
paper is ruled with a sufficient degree of accuracy to be 
used for most of the measurements necessary in a psy- 
chological laboratory. All circular measurements can 
be very much simplified if apparatus is constructed with 
such a diameter that a degree of arc is exactly equal in 
length to some one of the metric units. If the diameter 
of the circle is taken as 114.5 cm. a degree of arc will be 
very nearly equal to one centimeter. If this rule of con- 
struction in all circles is adhered to, the graduation and 
regraduation of the circumference of these circles will be 
a very simple matter, for degrees can be marked off on the 
circle with the aid of a flexible metric tape. 



The form of figure to be used in this exercise is repre- 
sented in the text of the Laboratory Manual on page 18. 
Some description is there given of the methods to be em- 
ployed in the preparation of the cards on which the 
figures are drawn. These cards should be large enough 
so that the boundaries of the cards shall not interfere with 
the inspection of the figure. A convenient size will be 
found to be 30 cm. by 15 cm. Any thin grade of card- 
board can be used, or a heavy grade of drawing paper will 
serve very well. The most convenient dimensions for the 
figures themselves are 10 cm. in length for Fig. A (Labor- 
atory Manual, p. 18); 3 cm. more or less, as suggested in 
the table of variations below, for the oblique lines; and 
angles from 15 to 75 degrees between the obliques and 
the long lines. The line B (Laboratory Manual, p. 18) 
should be drawn about 16 cm. in length, and its obliques 
should correspond to the obliques for Fig. A. A conven- 
ient method of describing these figures for purposes of 
record is by means of some such formula as the follow- 
ing: 10 cm., 3 cm., 45. 

The following variations may be suggested: 

10 cm. 

10 cm., 3 cm. 


Each member of the class should be required in some 



part of the exercise to work with figure 10 cm., 3 cm., 45. 
There will then be a basis for comparison of all observers. 

Modifications of a different kind from those suggested 
in the tables may be introduced by omitting certain of 
the oblique lines, in order to ascertain, for example, 
whether an upper oblique is more important for the 
illusion than the corresponding lower oblique. The 
length of the main line of the illusion may also be modi- 
fied. In that case the comparison between different 
figures should be made, not by direct comparison of the 
absolute amount of the illusion, but rather in percentages. 
Thus, if the illusion for 10 cm., amounts to 2 cm., the 
illusion will be stated as 20%. If the illusion when A is 
5 cm. in length amounts to 10 mm. it should again be 
stated at 20%. Further variations may be introduced 
into the figure by moving the whole figure to a greater 
or less distance from the observer; this reduces the size 
of the retinal image while maintaining a proportional re- 
lation between all of the different parts. Again, the 
figure may be turned so that its long lines are vertical or 

A convenient method of preparing these cards in such 
a way that the long lines can easily be matched with each 
other, is to set the cards against a straight edge and 
draw the long lines on both with one continuous ruling. 
In the same way, the easiest method of manipulating the 
cards is to place them in some form of tray or holder 
which has at one edge a straight strip. For this purpose 
a simple apparatus can be used consisting of a board 
that has tacked along one side a strip of wood which 
rises above the level of the board and gives an edge against 
which to rest the cards. This wooden tray can then be 
held before the observer, either in the hand or resting 
upon a table. 



The procedure of the experiment is now very simple 
as stated in the Manual. The observer should move the 
card on which is drawn figure A until the length of the 
line B seems to him to be equal to the line A. After he 
has adjusted the cards so that they are satisfactory, one 
of two methods may be adopted for the measurement of 
the results. First, two persons may work together on 
the exercise, one acting as observer and the other acting 
as experimenter, and measuring the results immediately 
after each setting. Secondly, the observer may mark off 

" F 

FIG. 9 

on strips of paper the length of B after each setting, and 
may then measure the marked strip after the whole series 
is finished. If the observer is allowed to measure the line 
B after each setting, he is likely to be more or less dis- 
tracted by his knowledge of the inequality between A 
and B. 

The apparatus for holding the cards and recording 
measurements may be made much more elaborate, as 
indicated in Fig. 9. A wooden frame 50 cm. long and 


30 cm. wide is made up of a wide board AA and a narrow 
board BB, held together by two side pieces shown in the 
figure. Between AA and BB slides the movable piece of 
board CC. This is matched to the two parts of the frame 
AA and BB, and is made loose enough to slide freely. 
It is held in position by a long, slightly bent strip of brass 
which acts as a spring and is placed between the upper 
edge of CC and the lower edge of BB. This spring will 
be found convenient in all types of wooden slides. It 
obviates the necessity of fitting the wood closely and frees 
the apparatus from all the effects of swelling through 
changes in humidity. Cards with the figure to be meas- 
ured can now be fastened with thumb tacks upon AA 
and CC. The card EE will in this case need to be some- 
what wider than described above. The advantage of 
the apparatus as thus far described is that the cards when 
once adjusted are held in position and can be rapidly 
adjusted without direct handling. Measurements which 
give no direct knowledge of the error can be made by the 
observer after each setting by measuring the amount by 
which C projects to the right beyond the frame. 

The remaining parts of the apparatus represented in 
Fig. 9 provide a means for recording the settings without 
direct measurement. At the extreme left of the figure is 
an arm on which is placed a spool of ticker tape. This is 
unwound, as shown in T, T, T, T, and drawn across the 
upper part of the frame BB. Two strips of metal seen 
near the ends of BB hold the tape in place. At the mid- 
dle of BB, and CC, are placed the recording points. The 
recording apparatus is constructed as follows : F is a short 
metal rod fastened to the sliding board CC and seen in 
the figure only at its end. The end of this rod F which is 
nearest to the ticker tape carries a small pin-point. The 
pin-point is held away from the paper by a coil spring 


placed around the rod. (For the details of such a spring 
see Fig. 18, page 29.) The observer, when he desires, 
can push against the upper end of the rod F, and thus 
drive the pin through the ticker tape, otherwise the pin 
is drawn back by the spring and the whole moves freely 
across the tape whenever the sliding board CC is moved. 
A second pin-point at G is fastened to a plate H , which is 
fastened in turn to BB. The pin-point at G is made 
double, so that its impression may easily be distinguished 
from the impressions of the point F. The plate H and 
the points at G are held away from the ticker tape by a 
spring. Pressure on the plate may, however, as in the 
case of the point of F, drive the pins at G into the tape. 
By means of a catch placed on F, a single pressure on F 
moves both F and the plate // downward to the tape 
against their respective springs, and there will thus be 
punctured in the tape two pin-records, one at G and one 
at F. Since G is fixed and F moves as the board CC moves, 
the distance between the pinholes in the tape will vary 
with every change in the positions of the two cards DD 
and EE. If the distance between the pins in G and 
F is determined when A and B are in reality equal, the 
amount of departure from this distance in the various 
settings will constitute a measure of the illusion. The 
advantage of this apparatus is that the observer can make 
a series of settings of the figure in rapid succession, and 
can readily record the results without the inconvenience 
of marking on strips of paper. 

A further refinement of the apparatus consists in at- 
taching to the sliding board CC a handle which will per- 
mit rapid coarse adjustments and slower fine adjustments. 
Such a handle is represented in Fig. 10. Let B represent 
the end of the board CC. The block K attached to B 
carries a screw rod D. The method of setting this rod 



into the block K is such that it shall have a screw move- 
ment and at the same time a hinge movement from D 
to D'. The upper end of the rod is turned so as to 
have a ball-shaped head carried on a shaft which is 
turned down to a diameter smaller than that of the ball 
or rod. This head is fitted into a cup which closes around 
it on all sides except one, where it is split so as to allow 
the narrow part of the shaft to swing forward into the 
position D' without drawing the shaft out of the cup. 
The bent brass strip at S acts as a spring to throw the rod 
into the position D against N. N is a half 
nut and is fastened to the frame of the ap- 
paratus. When the screw rod D is forced 
into this nut, the board B can be moved 
only by turning the rod and thus operating 
the screw in the nut. If the rod is drawn 
forward into the position Z)', the threads 
of the screw no longer engage the threads 
of the nut and the board B may be pushed 
freely back and forth. This makes possible 
a rapid, coarse adjustment. 

The whole apparatus can be clamped to a 
table at any convenient angle. It is better to 
use it in a position nearly vertical and at about 
the level of the eyes of a seated observer. 

An additional precaution which it is well to take in 
working with visual percepts is to cover all parts of the 
field not directly involved in the percept. A screen of 
gray cardboard with a circular or elliptical opening in the 
center, cut large enough to allow the lines to be seen, 
but not to expose any of the apparatus near the lines, is 
the most convenient means of avoiding distractions from 
the outlying visual objects. Such a screen can be tacked 
over the apparatus described above. 




A table showing the results of a series of measurements 
which emphasizes to an unusual degree the difference 
between the two types of setting, but is otherwise typical, 
may be given as follows : 

Table I 

Observer J. C. B. Figure, Miiller-Lyer; 10 cm., 3 cm., 45. Mch. 10. 
All measurements reported in millimeters. 

Illusion when 
B is too long 
at beginning. 

from avg. 
of five. 

Illusion when 
B is too short 
at beginning. 

from avg. 
of five. 

from gen. 







































Avg. 24.6 





General average for 10 determinations 20.2 M.V. 4.4 

A general table showing the results for various figures 
examined by the same observer is as follows: 

Table II 
Observer J. C. B. Date, March 10. 


Avg. 10 determinations 

M.V. ! 

10 cm., 1 cm., 45 



10 2 45 



10 3 45 



10 4 45 



10 5 u 45 





A table presenting the results for a group of observers 
using various figures is as follows: 

Table III 


10, 1, 45 

10, 3, 45 

10, 5, 45 

10, 3, 30 

10, 3,60 



M. V. 


M. V. 


M. V. 


















26 7 


The results reported in Table II are represented in the 
curve shown in Fig. 11. Similarly the results reported 




in Table III for the first three observers are given in the 
curves in Fig. 12. 


The first supplementary experiment deals with the 
Poggendorff illusion (Fig. 13). The lines A and B and 


FIG. 12 

the space between B and C should be drawn on one card ; 
the lines C and D on a second card. The two cards can 
now be adjusted on each other in the same way as the cards 
in the case of the Miiller-Lyer figure, until A and D seem 
to extend in the same direction. There is another method 
of treating this illusion; namely, by rotating one or both 



of the lines A and D until they seem, because of their ro- 
tation, to extend in the same direction. The method 
of measuring rotations is more complex than the method 
of sliding the cards back and forth over 
each other. The principle involved in the 
rotation method will be described later in 
a case in which it is clearly required, 
namely, in measuring the Zollner illusion. 
The second supplementary experiment 
deals with such an illusion as that shown 
in Fig. 14, where the point is in reality 
equally distant from the two circles, but seems to be 
nearer to the large circle. In attempting to draw the 
two parts of this figure on cards for purposes of adjust- 
ment in a way analogous to that described for the earlier 
figures, a difficulty which has not been serious up to this 
point makes itself emphatically felt. The line of division 
between the two cards forms in reality an additional object 
in the field of vision, and its presence can not be ignored 

FIG. 13 


FIQ. 14 

in trying to locate a point between two figures. The 
method must, accordingly, be modified in such a way as 
to make it possible to present two circles in the field of 


vision with an intermediate point which shall in no way 
be associated with a line of division between cards. 

The simplest method of meeting this demand is to draw 
the circles on a large card and then draw the point which 
is to be placed between them on the back of a glass which 
is large enough to cover the whole field. The glass with 
the point on its back surface may now be shifted over the 
fixed card until the point is in the desired position. The 
glass with the point may be fastened, if desired, to the ad- 
justable strip CC of the apparatus shown in Fig. 9 (p. 18). 
The glass may be fastened by boring holes through it and 
screwing it to the board, or by fastening along one edge 

I I I I I I I I I I 

vi 4 

FIG. 15 

either heavy gummed paper or electric tape, and fastening 
the paper or tape to the board. 

The method just described can easily be applied to the 
measurement of the illusions of interrupted space shown 
in Fig. 15. The distances XY and X'Y' of these figures 
are equal to YZ and Y'Z'. The point Z and the short 
line Z' may be drawn on the back of a glass, and may 
then be adjusted until XY seems equal to YZ, or X'Y' 
equal to Y'Z'. 

Another method of carrying out the experiments with 
circles of unequal size or filled and empty space consists 
in taking the observer into a dark room, and providing 
conditions of illumination which shall make it possible 


to manipulate the figures, without introducing any irrele- 
vant object whatsoever into the field of vision. This can 
be readily done as follows: Provide a box, one face of 
which consists of a milk glass which will distribute the 
light falling upon it as uniformly as possible. This milk 
glass should be illuminated by a light placed within the 
box. A black square or circle may be produced upon this 
surface by means of an opaque figure which will inter- 
cept the light supplied to the milk-glass surface from 
within the box. In the same way a dark point at any de- 
sired distance from the figure may be produced by a black 
point on the milk glass. A second figure of any shape 
or size desired can now be introduced on the back side of 
the milk glass, and can be 
moved nearer to the fixed 
point, or further from it, by 
means of a rod which extends 
directly into the box in such a 

way as not to cast a shadow 

i .11 i < FlQ - 

upon the milk-glass surface. 

The position of the adjustable figure can be measured 
either directly by means of a scale or by means of some 
recording device attached to the movable rod. This 
general method of escaping complicating objects in the 
field of vision by using illuminated surfaces and black 
objects can also be applied to other figures than the one 
in question. 

One of the illusions of direction resulting from lines 
crossing the main line of the figure is that which appears 
in the so-called Zollner pattern, Fig. 16. The apparatus 
for the measurement of this illusion is somewhat more 
complicated, because the direction of the line is to be 
measured and not its length. The simplest device b 
as follows: One of the long lines of the Zollner pattern 


with its obliques is drawn upon a card, and at a suitable 
distance from the end of the long line and at the same 
horizontal level as the lower end of the line, a black silk 
thread is fastened to the white card. This thread should 
be 114.5 cm. (or some fraction) in length, in order to facili- 
tate angular measurement. At a distance from the point 
of fastening which is equal to the length of the long line 
of the Zollner pattern, the thread should be made to pass 
behind a screen, which will thus cut off a suitable por- 
tion of the thread to be compared with the long line of the 
Zollner pattern. The thread should be kept under tension 

Fio. 17 

at its remote end by a weight or rubber band. Let the 
position of the thread be adjusted by the observer until its 
direction seems to him to be exactly parallel with the long 
line of the Zollner pattern. The deviation of the long 
thread from the position of true parallelism with the line 
of the Zollner pattern can then be read on a scale. The 
amount of this deviation constitutes a measure of the 

A piece of apparatus, which in a somewhat more elabo- 
rate form follows the principles already described, is shown 


in Fig. 17. An outer frame A A carries a piece of plate 
glass on the back of which is drawn a plain straight line. 
This line can easily be drawn on glass by means of an 
ordinary ruling pen and very thick India ink. The frame 
with the glass can be rotated, by means of the handle B, 
about a rod fastened to the frame at the back; and when 
the glass is so rotated it carries with it the long arm C. 
Held rigidly at the same center as that around which the 
glass and frame rotate is a wooden panel DD, to which is 
tacked any desired part of the Zollner figure. The fixed 
panel has connected with it the long arm E. The observer 
is required to rotate frame AA with the arm C about the 
fixed panel D, until the plain line and the long line of the 

Zollner patterns seem to him to be parallel; the degree 
of rotation is indicated much enlarged at the ends of the 
arms C and E. At the extreme ends of these arms are 
placed pin-points, which may be used to mark a paper strip 
and leave a permanent record of the relative positions of 
the arms. The device for making the pin impressions is 
represented in Fig. 18, and operates by means of the strings 
FF, which are carried to a point within easy reach of the 
observer who is seated before the frame and panel. The 
arm E carries at its end a wooden plate over which is 
drawn a strip of paper seen at P. Above and below the 
paper are pin-points which are held back from the paper 
by coil springs. The pins are driven into the paper 


whenever the strings at FF are pulled by the observer. 
If with this attachment the distance is recorded between 
the pin-points attached to the fixed arm E and the movable 
arm C, and if, further, the distance of the pin-points from 
the center of rotation is properly chosen, as for example 
114.5 cm., the apparent deflection of the line of the Zoll- 
ner pattern from its true position can easily be measured 
at once in centimeters and degrees of arc. 

For the remaining supplementary experiments it is 
only necessary to make measurements with one or the other 
of the methods above described, and to tabulate the re- 
sults fully as indicated in the table and curve given for 
such a practice series in the general introduction (page 



In all experiments dealing with monocular vision it is 
better to cover one eye with a loose shield rather than to 
bind the eye or require the observer to keep the eye closed 
by voluntary effort. A simple shield for the eye can be 
made of stiff cardboard covered with dark cloth. The 
shield should be large enough to cover the region from 
the eyebrows to the cheek bone, and from the bridge of 
the nose to the temple. It should be bent in its vertical 
axis and shaped at the sides so that the right and left 
edges will fit against the face as closely as possible. It 
can be held in position by means of a rubber band about 
the head. 

A second general requirement in monocular experi- 
ments is a head-rest. If table clamps and S-clamps and 
rods (page 9) are included in the general equipment 
of the laboratory, a simple and satisfactory head-rest 
can be set up without special additions to the general 
equipment. Let a horizontal rod be supported on a table 
by two upright rods at the level of a seated observer's 
upper teeth. Since the head naturally tends to drop for- 
ward, and since the upper teeth are firmly fixed in the skull, 
it follows that any support placed under the upper teeth 
will give the head a firm rest. Any other point of the 
skull which might be used for the same purpose has the 
disadvantage of being covered with loose skin which per- 
mits more or less movement of the head against the rest. 
If it is desirable that the head be brought to the same 




M S 

position in a number of successive experiments, or if it 
is necessary, as in this exercise, to hold the head in the same 
position for a long time, the fixation of the teeth can be 
made much more complete by covering the supporting 
bar with a layer of sealing-wax into which the teeth may 
be pressed while the wax is warm. The mold of the 
teeth thus produced upon the bar can be used to bring 

the head back into its original 
position even after the ob- 
server has moved away for a 
time. Instead of the hori- 
zontal rods just described, 
a wooden strip small enough 
to be taken between the teeth 
can be supported by clamps 
from the table or from a chair, 
and the upper teeth can rest 
upon this wooden strip. 

It is frequently desirable to 
have a portable head-rest 
which can be set up in smaller 
space than the table-rest de- 
scribed. Fig. 19 represents 
a very convenient form of 
general head-rest. A piece of 
iron pipe P is screwed or 
driven into a heavy tripod base B. If it is desired, a 
number of pipes of different lengths may be provided for 
various purposes. A block of wood W, split into halves and 
having in its center a hole somewhat smaller than the 
iron pipe, can be clamped firmly to the pipe P by 
means of screws RR, which pass through the block and 
are provided with thumb nuts on one side. This 
block can be loosened by unscrewing RR, and can then 

FIG. 19 


be set at any desired level up and down on P. At 
the sides of the block are fastened adjustable strips of 
iron as indicated in detail at S, T, M. S is a thumb- 
screw which clamps the strip T against the block. By 
means of the slot M in the strip, the strip may be made 
to project to any desired distance beyond the block. 
This arrangement also makes it possible to set the strip at 
any desired angle, as at B. At the outer end of the strip 
is a device for clamping a cross rod. This device con- 
sists in a slot N in the strip T, and a bent piece of iron 
F, with a double slot corresponding to the slot N in the 
long strip. The bent piece F has a screw G passing 
through its middle. When now the bent piece is slipped 
over the strip, the three slots give an opening which may 
be regulated in size by means of the screw G. If this 
screw is set down against the end of the strip T, the open- 
ing left between the three slots can be reduced in size so 
as to hold a rod firmly in position. A second adjustable 
iron strip K is fastened on the opposite side of the wooden 
block. Between these two strips a bamboo rod can be 
fastened in the slots N N', and on the rod a sealing-wax 
impression for the teeth can be made as described above. 
The rest can be used in two positions. Either the back of 
the head can be placed against the block and the rod 
with the teeth-rest be adjusted to hold the head back 
firmly against the block, or the iron strips can be set up 
from the wooden block, as in B, the block acting in this 
case as a standard on which to fasten the teeth-rest. In 
position T the space in front of the eyes is left entirely 
free; in position B the space around the ears is also left 

A perimeter, which is the most convenient device for 
measuring positions on the retina and stimulating its 
different areas, can be constructed in very simple form 


from a child's hoop. The hoop should be as large as 
possible; it should be cut in two so as to give a half circle, 
and it should be graduated in degrees. The easiest 
method of graduating the hoop is to determine by measure- 
ment the total circumference of the hoop and then, after 
dividing this by 360, to mark off by direct measurement 
every -jj-^- of the half circumference. The half hoop 
should be fastened on some kind of firm upright stand- 
ard, if necessary on the wall. Other points of fastening 
may be suggested. Thus, a heavy rod fastened to a table 
is a very good holder and gives the experimenter better 
opportunity of getting behind the perimeter. Even 
better is a special base similar to that described in connec- 
tion with the head-rest. The point of fastening should 
be on the same horizontal level as the observer's eye. It 
is often easier to adjust the height of the observer than of 
the apparatus. The observer should be placed with one 
eye at the center of the circle of which the hoop forms the 
circumference; the other eye should be covered; and the 
observer should be required to look steadily at the point 
on the semicircular hoop at which it is fastened to the 
standard. This center should be marked with a spot of 
white paint in order to make it easy of fixation. The 
position of the perimeter will determine the part of the 
retina which is to be examined. The most convenient 
position is that in which the perimeter is placed in the 
horizontal plane. It may be placed vertically or obliquely. 
If the whole retina is to be examined, it must be placed 
successively in a number of these positions. 

A small disk of colored paper two or three centimeters 
in diameter should be attached to a rod and should be 
held successively at various positions along the graduated 
circumference of the hoop. It is better that the color be 
exposed for a brief interval and then withdrawn, rather 


than exposed continuously, as the retina fatigues very 
easily, especially in its eccentric parts. The background 
against which the colored disks appear can be made 
either gray or black. It should, however, be uniform in 
its color tone. In certain elaborate experiments with 
the perimeter, uniformity of background is secured by 
going into a dark room and exposing the various colored 
papers by means of some artificial light which can be 
turned off when the apparatus is not in use. When the 
observer is thus taken into a dark room the condition of 
the retina is modified. It becomes dark-adapted and its 
sensitivity to colors is completely modified. The investi- 
gation when conducted in ordinary daylight gives results 
which are appropriate to an eye which is light-adapted. 
The disadvantage of trying an ordinary class experi- 
ment in a dark room is that dark-adaptation is not easy 
either to secure or maintain. In order that it should be 
complete, the observer must sit in a dark room for at least 
fifteen minutes and possibly more before the beginning of 
the experiment. Furthermore, every time the light is 
turned on there is a tendency to return to the condition 
of light-adaptation. The experiment must, therefore, 
proceed very deliberately. If the experiment is con- 
ducted in ordinary daylight, the background should not 
be so bright as to be dazzling, and this end is best secured 
by hanging about the perimeter either a gray paper 
or gray cloth. The color then appears on a uniform 

A simple device which is easy to set up but complex 
in its measurements, consists in moving the eye rather 
than moving the object along the circumference of a circle. 
In this case the center where colors are exposed consists 
of a small circular opening in a cardboard screen. The 
eye can either fixate this opening directly or can look 


at some point at the left or right of this center. The dis- 
tances to the right and left of the center of fixation being 
measured directly as straight lines, a trigonometric cal- 
culation is required to reduce the measurements to de- 
grees. The simplest method of reduction is to fix an eye- 
rest consisting of a small ring at a known distance from 
the point where the color is exposed, and then after cal- 
culating the tangents for various degrees of circular move- 
ment, mark the scale to the right and left of the opening 
through which the color is exposed in linear extents which 
correspond directly to degrees. Such a graduated scale 
being prepared, the eye is turned say to a position 80 
from the center at which the color is exposed; the color 
may now be exposed to the eye, and since the eye has ro- 
tated outward through 80, the light will fall on the ret- 
ina at a point 80 from the center of vision. The method 
of exposing the colored paper in this case consists in merely 
uncovering either a piece of colored paper or a source of 
colored light. The screen which has covered the colored 
light up to the moment of exposure should be of the same 
intensity of gray as the general background, which in this 
case consists of the screen upon which the graduations 
for the eye movement are placed. This form of appa- 
ratus, which is known as a campimeter, has the advantage 
over the perimeter that all sorts of colors can easily be 
placed under the opening in the screen. Color mixers 
of different forms, to which reference will be made in 
the description of the apparatus for the next exercise, can 
easily be used in connection with the campimeter, while 
they can not be used with any simple perimeter. 

Colored papers for this experiment and the next can be 
secured from C. H. Stocking Co., 39 West Randolph 
Street, Chicago, 111., who are the American agents for 
Rothe, the mechanic of Professor Hering, of Leipzig. 


Several systems of papers have been prepared by Ameri- 
can manufacturers, notably those prepared by Milton- 
Bradley and the Prang Co., but these are inferior for 
purposes of scientific investigation to the selected Ger- 
man papers referred to above. Various devices for 
using transmitted rather than reflected light have been 
prepared. The most successful of these consists in col- 
ored gelatines, which are also of foreign manufacture 
and are to be had of Stoelting Co. A number of gela- 
tines placed together can be so combined as to pro- 
duce colors that are approximately pure. They are, 
however, by no means as bright as reflected colors from 
the Hering papers. Methods for the determination of 
the purity of colored fields and methods for the deter- 
mination of the intensity of colored fields are unnecessary 
for anything except the more elaborate tests of the sort 
suggested in this exercise. 

If it is desired to make these tests, the ordinary spectro- 
scope methods employed by the physicists may be used 
for finding the quality of the colors. The establishment 
of the equations of like intensity of colors is of somewhat 
greater importance, and a simple method can be employed 
as follows: Two colors which are to be compared with 
reference to their intensity are brought into such a position 
that they stimulate the extreme periphery of the retina. 
At this part of the retina their color qualities are entirely 
lost and they are seen as gray. Their relative intensities 
can now be compared so as to determine which is the 
brighter. It is ordinarily not possible to change the 
character of the papers so as to make them equal in 
intensity, but the results of the equating of the color 
intensities can be recognized in evaluating the results of 
experiments with the papers. 

One of the most common sources of difficulty in this 


experiment consists in the inability of the ordinary ob- 
server to describe with accuracy his color experiences. 
Slight differences in color qualities are not covered by the 
vocabulary of many observers. As a preliminary to this 
exercise it is therefore advantageous to spend some time 
exhibiting a series of color qualities and preparing a series 
of designations for them. For this purpose an excellent 
series of color specimens and color terminology will be 

found in the Standard 
Dictionary under the 
term spectrum. Oth- 
erwise the observer 
should be carefully 
instructed to follow 
the recommendation 
given in the Laboratory 
Manual of describing 
each succeeding color 
experienced in terms 
of its differences from 
the color which im- 
mediately preceded it. 
Observers can very 
frequently tell the 
direction of a color 
change when they have no adequate single phrase with 
which to designate the color itself. 

More elaborate forms of apparatus than those described 
thus far have been devised and are available for purposes 
of the experiment here under discussion. Perimeters 
are used by oculists for the determination of the retinal 
field without reference to the modification in the character 
of color vision. Elaborate frames carrying different de- 
vices for holding and recording the position of stimuli 

Fio. 20 
From the catalogue of Meyrowiti 


have been prepared. Several of the more complete forms 
of automatic recorders may be found by consulting the 
catalogue of E. B. Meyrowitz, 104 East 23d Street, New 
York City. (See Fig. 20, which shows one of these.) 

The metallic half circle here used is supported on a heavy 
base. At its center the half circle may be rotated through 
every angle of the complete sphere. Its angular position 
is measured by a disk shown in the figure. This disk 
carries a printed chart of the retina, marked off in degrees. 
An automatic recorder indicates on this chart the posi- 
tion of any source of light which moves along the half 
circle. The observer, with his head fixed in the rest 
shown at the right of the figure, is seated in such a posi- 
tion that the center of his eye is at the center of the cir- 
cle of which the half circle is a part, and in the same 
horizontal plane as the center of the perimeter. 

It is desirable, as noted above, that the color which is 
to be exposed shall not be constantly visible to the ob- 
server. Various devices may be suggested, such as cov- 
ering the color when it is being adjusted. Still better is 
the construction of a perimeter somewhat simpler than that 
shown in Fig. 20, but allowing the attachment of the light 
box represented in Fig. 21. A box shown in outline at 
AAA is painted black on the outside, so that it shall offer 
no positive stimulus. It is supplied inside, as shown at E, 
with an incandescent lamp. Some lamp which gives ap- 
proximately white light, such as the tautalum lamp, is 
preferable to the ordinary incandescent lamp. The box 
is open at the point F, there being at this point a small 
circular opening 2 cm. in diameter. Just behind this open- 
ing and in such a position that it reflects the light from 
the lamp E is placed a colored paper C. Any color may 
be inserted at the point C, and when the box is closed and 
the lamp turned off the color will be invisible to the ob- 



server. By means of a simple switch at S the lamp E 
may be turned on, when the color will immediately be 
visible. If an ordinary incandescent lamp is used at E 
an additional feature must be introduced into the box 
because the color of the light from such a lamp is decidedly 
red. There must be introduced at RR a layer of colored 
gelatine which will modify the light so as to make the 
light less decidedly red. A combination of green and blue 
gelatine can be secured which, held between two sheets 

FIG. 21 

of glass in the position RR, will modify the light from E 
before it reaches the colored surface C, and give satis- 
factory color effects with all papers. 

The perimeter and its attachments may be of any size 
desired. The larger the circumference, the easier will be 
the readings for different positions of the color. Further- 
more, if the base and circle are made heavy enough they 
may carry any desired additional apparatus, such as color 
mixers. In the larger forms a quarter circle counter- 



balanced at one end by a heavy weight serves better than 
a full half-circle. 


The following table shows a very good series of observa- 
tions. In general the observations are less consistent. 

Observer A. B. C. Right Eye. Colors from Hering Papers. 
October 8. 







Light, without 

Light, no color 




Light, no color 

Light, without 

Light, no color 





Possibly red 

Orange, pale 

Yellow, distinct 

Pale yellow 


Yellow, dis- 

Yellow, pale 

Yellow, no 

Blue, dis- 





Yellow, toward 


Yellow, no 

Blue, paler 





Yellow, pale 






Red, distinct 

Very pale, pos- 

Yellow, same 

Blue, very 

sibly green 

as last 



Red, distinct 


Yellow, same 







Blue, same 







Blue, same 

Fig. 22 shows graphically the results of a series of such 
observations for a class. The letters represent the vari- 
ous observers. The general result is here made obvious 
and agrees with the special table reported above. It will 
be seen from the figure that not all observers show so 
marked a difference as is shown in the table between red 
and green on one hand, and yellow and blue on the other. 



Among the additional problems described under Exer- 
cise II, is one for plotting the projection of the blind-spot. 
No elaborate apparatus is necessary for this experiment. 
Any means of holding the head in a fixed position before 
a large white surface will satisfy the requirements. A 
block of wood 25 cm. in length, placed upon a table, fur- 
nishes the simplest head-rest. If the teeth rest against 
the upper end of the block, and the eye is allowed to fixate 

0o ; o 20 30 40 60 60 70 80* 90 

fiea .-_____ 
8. M. J. H. f. Q. T. Qr. To. Me. 


F. T. Me. J. S. H. C. Q. M. Qt. Ta. 

C. T. H.J.S.Ta. O.F.Mc. M.Q. 


Me. T.8. C.Ta.6. Qr.F.J. H.M. 

Fio. 22 

a point on a sheet of paper lying flat on the table under 
the block, the eye can be held nearly enough in one posi- 
tion to permit the experiment to go forward. If the right 
eye is used, the point of fixation should be somewhat left 
of the center of the paper, and conversely, if the left eye is 
to be used, the point of fixation should be on the right. 
After the head has been brought into position and the eye 
fixed upon the point drawn on the paper, a strip of paper 
with a black spot at the end should be brought into the 


field of vision and should be moved about until the spot 
can not be seen. Evidently the spot is now within the 
area which casts its image on the blind-spot. In order to 
determine the boundary of this area the spot should be 
slowly moved to the right or left until it can just be seen 
by the observer. At the point where it becomes visible 
after being in the blind-spot, it is evidently emerging from 
the projection of the blind-spot and the fact should be 
recorded on the paper by means of a point. The movable 
strip of paper is now drawn back so that the spot falls 
in the blind area again; it is again moved gradually for- 
ward until it appears at some other point on the boundary 
of the blind area where a second point is marked. In 
this fashion a sufficient number of points may be definitely 
described on the edges of the area to indicate the outline 
of the projection of the blind-spot. Two cautions are 
necessary. It is better to determine the boundaries of the 
projection by drawing the point out of the area than by 
moving it into the blind area from some part of the field 
where it is visible. If the point is moved into the area 
from the parts of the field where it is visible, its movement 
can be seen and there is a strong natural tendency to fol- 
low its movement. Secondly, the eye must be steadily 
fixed on the point marked on the paper, otherwise the out- 
line of the blind area will not be determined with exactness, 
for with every movement of the eye the position of the blind- 
spot changes. 

If a convenient head-rest is at hand, such, for example, 
as that described on page 32, it should be used instead 
of the block above described, and the observer should be 
allowed to look at a large sheet of paper which has been 
hung vertically in front of the face. The further procedure 
is exactly the same as in the first case. 

The second supplementary experiment under this ex- 


ercise requires that the degree of apparent curvature of a 
straight line at the periphery of the field of vision be 
measured. As in the earlier parts of the experiment, let 
the head be held in a fixed position and the eye fixated 
upon a determined point. Arrange on the edge of the field 
of vision a series of points in such a relation that they 
will seem to the observer to extend in a straight vertical 
line. They will, in reality, describe an arc rather than a 
straight line, and the degree of the curvature can be readily 

The third supplementary exercise suggested on page 
25 of the Laboratory Manual can readily be set up by 
drawing on white cards black geometrical figures. The 
figures should have a general diameter of about 3 cm., and 
should be exposed on a perimeter or with the aid of a 
campimeter, as indicated in the experiments with colors. 
If the results for these figures are tabulated and compared 
with the results for the various colors, it will be, found that 
there are no intermediate zones of partial sensitivity for 
these figures. It will also be found that only the central 
parts of the retina which are highly enough developed to 
give complete color differentiation are highly enough de- 
veloped to make possible the recognition of forms. 

To determine the sensitivity of different parts of the 
retina for slight changes in illumination, prepare series of 
cards with gray circles of different intensities on a black 
background. Graded grays for this purpose can be ob- 
tained from Stocking Co. The first card of this series 
should show very little contrast, the second more, the 
third still more, and so on. By successive trials one card 
can be found for each part of the retina which gives the 
just recognizable contrast. The procedure here is similar 
to that followed in exposing colors to the different parts 
of the retina. Beginning with a very slight contrast, ex- 


pose it to various parts of the retina and determine where, 
if at all, it can be recognized. After this increase the 
contrast and again expose to different parts of the retina. 

Another method of determining the sensitivity of differ- 
ent parts of the retina is to measure the amount of change 
in the intensity of light which is necessary to arouse in 
the observer the recognition of the fact that the intensity 
is changing. The observer may be allowed first to look 
at an area that is illuminated from behind by means of a 
certain source of light. If now the source of light is 
moved steadily away, the intensity of its illumination will 
be decreased in the well-known physical ratio according 
to which the intensity of a given light is inversely propor- 
tional to the square of its distance. If, therefore, we 
measure the distance through which such a light must be 
moved in order to give a just noticeable increase or de- 
crease in the illumination, we have a measure for the sen- 
sitivity of that part of the retina. Care must be taken in 
such an experiment as this to move the light at a uniform 
rate. The experiment can be carried out most easily in 
a dark room. It is also advantageous where possible to in- 
troduce into the field of vision two illuminated areas, one 
of which is kept at the original intensity and the second 
one of which is modified in the manner just described. 
Such determinations may be carried out for different parts 
of the retina. The method here suggested is the regular 
method of photometry and will be described again in 
another connection. (See page 131.) 

Experiments of this type give different results when the 
eye is dark-adapted than when it is light-adapted. 

If the experiment is to be made for the dark-adapted 
eye and a dark room is not available, a substitute for the 
dark room can be provided by constructing a large box 
which can be made so as to exclude light except at the points 


where it is admitted for experimental purposes. Such a 
box should be painted black inside and should be pro- 
vided at one end with a hood which can be thrown over 
the observer's head so as to exclude all light from that end. 
At the end of the box opposite the observer a screen may 
be provided which is perforated with any desired number 
of openings. The intensity of illumination which is al- 
lowed to come through these openings can easily be regu- 
lated. One convenient method of reducing the light is 
to hold in front of the opening a number of sheets of thin 
paper. The number of sheets of paper necessary to pro- 
duce a just perceptible difference in the illumination of 
two neighboring points furnishes a rough but convenient 
method of measuring that difference. A more exact 
method is to interpose a series of plates of milk glass be- 
tween the source of light and the observer, and to control 
the area of each of these glasses through which light may 
pass by means of an adjustable diaphragm such as that 
shown in Fig. 33, page 67. The successive plates of 
glass and their diaphragms may be enclosed in a box so 
as to secure light from one source only. 

In all of these experiments it is highly important that 
fatigue and after-images should be avoided. These can 
be avoided by allowing sufficient intervals to elapse be- 
tween successive experiments. 

The last supplementary experiment deals with the 
matter of adaptation to light and darkness and requires 
no new methods. 



The best introduction to this exercise consists in showing 
the observer a solar spectrum in which the different 
colors are exhibited in their natural intensity and order. 
Such a solar spectrum can be produced by allowing a 
pencil of sunlight to fall upon an ordinary prism. The 
rays coming from a prism may be projected upon a screen 
or may be examined directly by the observer. Still better, 
let the observer look into a spectroscope, which has the 
advantage over the prism of magnifying very much the 
color areas and of rendering the different qualities much 
more distinct. 

For purposes of experiments with color, the light from a 
solar spectrum is very difficult indeed to control, and, ex- 
cept with the most elaborate equipment, it can not be used. 
For this reason recourse is commonly had to rays of light 
reflected from colored paper surfaces. The light which 
comes from colored surfaces can be controlled within wide 
limits, both in quantity and in position, and can easily be 
mixed. No special materials or methods beyond those 
described in connection with the last exercise are necessary 
to make possible the first part of the experiment. 

The first and simplest method of mixture is to utilize 
such a reflection apparatus as is represented in Fig. 23. 
This apparatus consists of two boards placed at right 
angles to each other. Between these two boards is set a 
plate of plain glass which rests in the angle between the 
boards without fastening at the bottom. The glass can be 



supported by a string passing over the top of the vertical 
board and can be set at any desired angle by adjusting 
the length of the string. The glass should be of good 
quality so that it will have a uniform surface and will be 
as nearly transparent as possible. The observer should 
look down upon this plate of glass from above, as indi- 
cated by the diagrammatic eye, 0, in the figure. If a 
sheet of colored paper is placed on the lower board on the 

Fio. 23 

line OA, rays of light from this colored surface will pass 
upward directly through the plate of glass into the ob- 
server's eye. A second colored surface placed on the 
vertical board in the line OB will send its rays to the plate 
of glass in a horizontal direction, and a part of these rays 
will be reflected from the upper surface of the glass and 
will enter the eye of the observer along a path which co- 
incides with that followed by the rays from the colored 



surface on the horizontal board. The intensity of the 
light along the line B can be modified by changing the 
angle of the glass. By this means any desired colors 
can easily be mixed. The limitation 
of this method is that it does not per- 
mit of easy quantitative determinations. 
Furthermore, it does not permit a 
mixture of more than two colors at a 

A more common method of mixing 
colors consists of some system by which 
disks of colored paper can be locked together and rapidly 
rotated. The simplest device of this sort is a color-top 
(Fig. 24). A simple wooden shaft and tightly fitting ring 
hold in position small disks of colored paper. The whole 
is rapidly rotated by the fingers or by some string device, 
with the result that the colors, which are fastened to the 
top, pass so rapidly in succession before the eye that the 

FIG. 24 

FIG. 25 

effect produced by the different colored surfaces is prac- 
tically continuous. The disks used with the top and other 
rotating devices have small circular openings in the center 


and are cut along one radius. Fig. 25, A, represents two 
such disks being slipped over each other so as to leave a 
part of each disk exposed. The final relation of the disks 
is represented in Fig. 25, B. Obviously the relation can 
be varied at will so as to expose as much as is desired of 
either disk. In like manner, several disks may be placed 
together. Furthermore, it will be easy to measure the 
amount of each colored surface exposed. For this pur- 
pose a protractor is placed over the disks and the area 
of each sector is recorded in degrees. A transparent 
celluloid protractor is a very convenient form for this 

Colored papers suitable for this experiment have already 
been discussed on pages 36 and 37. If the papers are pur- 
chased for use in laboratories of moderate equipment, it 
will generally be found advantageous to purchase the pre- 
pared disks. If the papers are to be extensively used in a 
great variety of experiments, it is desirable to provide a 
cutter which will make it possible to prepare disks from 
large sheets of paper. A convenient cutter may be briefly 
described as follows: A metallic bar has at one end a 
handle and a sharp point. The point is driven through 
the paper into a board underneath; it now serves as a cen- 
ter of rotation for the cutter. On the bar is fastened a 
knife which can be adjusted by sliding it along the bar 
to any desired distance from the center of rotation. This 
knife is pressed against the paper and the disk cut out 
by rotating the knife so as to describe the circumference 
of a circle. 

The color-tops described offer such small surfaces to 
the view of the observer that contrast and other distract- 
ing effects impair seriously the value of the measurements. 
A mixing device which utilizes colored disks of larger size 
made of cardboard, is included in Milton-Bradley's Pseu- 



doptics.* This mixer takes advantage of a principle 
which is employed in a simple child 's toy. Two holes 
pass through a small metal disk on opposite sides of the 
center of the disk. Through these two holes are drawn 
two strings. If now the extremities of these strings are 
held firmly in the fingers and the string is twisted about 
itself, the disk can be set in rapid motion by pulling at 
the ends of the string. Such momentum can be attained 
by the disk in its rotation that it will twist the string 
around itself in the opposite direction to that in which it 
was twisted at first, provided the operator 
relaxes the tension on the string after the mo- 
mentum is well established. After the disk 
has wound the string around itself it will 
come to rest. It can now be set in motion in 
the opposite direction by again pulling on 
the ends of the string. The disk can be 
kept in successive rotation in one direction 
and the other by simply drawing and relax- 
ing the ends of the string. The central per- 
forated metallic disk through which the 
string passes is made in two parts in the 
color mixer, the two parts being screwed to- 
gether. Between these two parts are clamped 
the colored disks to be used in the mixing. 
Pasteboard disks must be used in this case rather than 
paper, otherwise the successive rotation in opposite direc- 
tions would result in air currents getting between the 
disks and tearing them. The mixer is shown in section in 
Fig. 26. 

*The Milton-Bradley Co., Springfield, Mass., issue a collection of charts and 
simple experimental devices for the illustration of a number of visual phenom- 
ena. It was announced, after this collection of charts and experiments had been 
in use some time, that it was devised by Prof. Munsterberg. The collection is 
extremely valuable as a means of demonstration and laboratory experiment. 
It should be in the hands of every teacher of psychology. The price of the 
three boxes is five dollars. 

FIG. 26 


A third form of color mixer which utilizes colored disks 
consists of a combination of wheels driven by the hand. 
In order to secure the proper speed for mixing colors 

with such apparatus, it is 
necessary to have a series 
of cogs or belts between 
the handle and the rotat- 
ing shaft on which the 
disks are held. A satis- 
factory mixer of this type 
is little, if any, less ex- 
Fia - s 7 pensive than an electric 

From the catalopje^Rothe, Leipzig. motor t(> ^ described in 

the next paragraph. It 

should not be purchased unless it is quite impossible to 
secure proper electric connections for the motor. A good 
form of hand mixer made by Rothe in Leipzig is shown 
in Fig. 27. Most of the other hand mixers are unsatis- 

A fourth device which is very satisfactory, consists in 
rotating the colored disks by means of an electric motor. 
An ordinary fan motor gives the rate necessary. The shaft 
of such a motor is supplied with an attachment, between 
the two parts of which the disks can be clamped. This 
addition to the shaft of the motor is known as an arbor 
and is the same in principle as the disk holder shown in 
Fig. 26. Care must be taken to place the disks in such a 
relation that when the motor rotates the air currents 
will not spread the disks apart. If paper disks are used, 
an uncut disk of cardboard should be placed behind the 
colored disks as a backing disk. 

Some fan motors when once set in motion are likely to 
continue rotating because of inertia for an inconveniently 
long period of time. It is, therefore, desirable to pro- 


vide some sort of brake for these motors. The most con- 
venient brake consists in an electrical adjustment by means 
of which the current, which has been passing through the 
coils of the field magnet and the armature, may be made 
to pass through the coils of the field magnet only. By 
such a device the field magnet will act as a magnet and 
will attract the armature so as to bring it to a standstill. 
The connections required are represented in Fig. 28. 
When the key at K is open, the current passes through 
the series of connections 1,A,Z,F,3, F f ', 4, that is, through 
the field and armature both. When K is closed, the cur- 
rent passes through the circuit XI, X2, K, X3, F, 3, 
F f , 4, that is, through the field magnets only. 

The method of procedure with the motor color mixer is 
to clamp in the arbor two sets of disks, one of large size, 
one of small size. The larger outer set may be, for ex- 
ample, in the first equation, red and yellow, the inner set 
should then be orange, white, and black. Or the reverse 
relation may be chosen by using small red and yellow 
disks and large disks of 
orange, white, and black. 
An arbitrary combination of 
such disks should be rotated, 
and then in successive trials 
the correct relation should be 
worked out by readjusting 
the disks until the same color 
appears in the inner and 
outer areas. 

There are certain devices 
by means of which the re- 
lation between disks can be changed while the^ are in rota- 
tion. These do not permit of fusion of two concentric 
sets of disks, the matching must accordingly be between 



two color surfaces placed side by side. The great ad- 
vantage of changing the disks during rotation is that 
the effect of each change is immediately seen without 
stopping for the slow readjustment and trial necessary in 
ordinary mixers. The difficulty in constructing such a 
piece of apparatus as this is that there must be provision 
for the rotation of the disks, and at the same time there 
must be some fixed portion of the apparatus which can be 
easily moved by the experimenter so as to change the 
amount of exposure of the disks. A device constructed 
by Marbe which has these parts is shown in Fig. 29. It 

Fio. 29 
From the catalogue of Zimmerman n, of Leipzig, Germany 

consists, first, of a screw which can adjust the length of a 
long catgut violin string. The string is fastened to this 
screw by means of a swivel joint which makes it easily 
possible for the catgut to rotate, while at the same time it 
can be drawn backward and forward by turning the screw. 
The catgut passes forward from this screw to the center 
of a hard rubber disk. Here it passes over a pulley and 
turns in a direction perpendicular to its first direction and 
passes outward to the rim of the disk. It is now turned 
on a second pulley and passes from this point over a num- 
ber of pulleys describing a circle around the back of the 
disk. By passing over these pulleys, it describes a circle 


which has its center at the point of its first contact with the 
disk. After passing around the full circumference of this 
circle, the catgut is attached to a hard rubber ring which 
is held in position against the tension of the catgut by a 
spring. The hard rubber ring is capable of moving 
around the disk under the tension of the catgut in a clock- 
wise direction, while by virtue of the spring it tends to 
return to its original position by moving counter-clock- 
wise. This hard rubber ring has fastened on the surface 
opposite to that on which it is connected with the catgut, 
a colored disk, such as has been described in connection 
with the color-mixing motor. A second color disk is 
fastened firmly in position against the hard rubber disk 
around which the hard rubber ring is capable of rotating. 
The two color disks are interlocked along radii in the 
manner illustrated in Fig. 25. We have thus provided 
two colored disks, one of which is capable of being changed 
in its position by means of the catgut, the other of which 
is independent of the catgut. If now the screw which con- 
trols the length of the catgut is turned back, the catgut 
will draw the adjustable ring in a clockwise direction and 
will change the relation of the two colored disks to each 
other. If the screw releases the catgut somewhat, the 
adjustable ring will be moved by its spring in a counter- 
clockwise direction and will reverse the relation between 
the two colored disks. The disk and ring to which the 
colored papers are attached are mounted in bearings 
and are made to rotate by means of a belt connected with 
an electric motor. The movement of this disk and ring 
with the colored papers involves a rotation of the catgut 
also, but this was provided for in the swivel joint described 
above. This complex apparatus is hardly necessary for 
ordinary laboratory work. 



The following table presents in number of degrees the 
results of the mixtures required in the exercise: 

169 Red + 191 Green =111 Yellow +210 Black + 39 White 

153 Red + 207 Gr. Blue = 228 Black + 132 White 

205 Red + 155 Blue = 358 Purple + 1 Black + 1 White 

96 Red + 162 Green -I- 102 Blue = 253 Black + 107 White 
138 Yellow + 168 Blue + 64 Green = 247 Black -I- 113 White 

71 Orange + 127 Blue + 162 Green = 248 Black + 112 White 

These equations are valid, of course, only for the par- 
ticular colored papers used and for the general illumina- 
tion under which the conclusions were worked out. The 
papers used in this case were the older Hering tissue 
papers, and the illumination was that of a large general 
laboratory in which the light was admitted from large 
windows about three sides of the room. 

A convenient graphic device for representing the re- 
sults of mixtures from a group of observers is represented 
in Fig. 30. The areas at the left represent, in each case, 
the relative amounts of colors necessary to match com- 
binations of gray with the intermediate colors repre- 
sented on the right. The areas given were secured by 
averaging the results obtained from a class of fifteen 
students. By comparing the various figures for the succes- 
sive combinations with each other, it will be seen that 
the general law of color mixing is clearly brought out. 

It will sometimes occur in a laboratory class that some 
member will exhibit a form of color-blindness which will 
appear from the abnormal character of his equations. 
Tests for such color-blindness have been carefully pre- 
pared, and it is better to use one of the regular tests than 
merely to try rough experiments. A rough test may, 
indeed, be made by collecting a series of commercial 
wools of different dyes. These being placed together in 



an unassorted mass, the observer is requested to take out 
those which seem to him to match even remotely some 
vivid green or red. If he successfully passes the test for 
red and green it is probable that he is not color-blind. 
The use of names for the color-blind test is very mislead- 
ing, for many persons who have a clear recognition of the 
different simple colors are unable to give the names of 
these colors, while, on the other hand, persons who have 

























FIG. 30 

some defect in color vision have learned enough of the 
relation between various color names and the experiences 
which they have derived from the different colored lights 
to stand a test even better than some normal individuals. 
The special tests are worked out by putting together a 
number of confusion colors which to the normal eye are 
clearly distinguishable, but to the abnormal eye seem 
alike. One of the best of these tests is that prepared by 


the physiologist, Nagel, who is himself color-blind. The 
cards may be had by addressing Professor Nagel, Uni- 
versity of Berlin, Berlin, Germany, or will be supplied by 
the Yale Psychological Laboratory. 


The first supplementary experiment deals with the 
facts generally classified as light contrast and color contrast. 
To get a good illustration choose some complex color, such 
as a dark reddish brown. Place a small piece of paper thus 
colored on a bright red background and another piece 
with exactly the same color on a dark green background. 
On the red ground it will seem very dark and dull in color. 
On the green it will seem lighter and decidedly more red 
in quality. 

In general the experiments in light and color contrast 
are most effective when the background is large and the 
contrasting color small. Furthermore, the contrast is 
most effective when the general intensity of illumination 
of the fields is not great. The desired reduction of illu- 
mination can be secured by covering the contrasted papers 
with thin tissue paper. Contrast effects appear, finally, 
most clearly when shades of little saturation are placed 
upon vivid backgrounds. One of the boxes in Milton- 
Bradley 's Pseudoptics contains various gray and colored 
rings especially adapted to the demonstration of contrasts. 
Professor Witmer's Analytic Psychology, published by 
Ginn & Co. also exhibits contrasts in very striking forms. 

A very good way of demonstrating contrasts is to set 
up a piece of colored glass so that the light from a window 
shall fall through it on a large white sheet of paper. Be- 
tween the colored glass and the sheet of paper place some 
object which will cut off the light from the glass, thus 


leaving a part of the sheet of paper illuminated by the 
general diffuse light of the room. This uncolored area on 
the paper will seem by contrast with the general field in 
which it lies to be colored in the shade complementary 
to the background. 

A very convenient method of demonstrating light con- 
trasts is to prepare a disk such as is shown in Fig. 31, A. 
This disk, it will be seen, has around the center a complete 
circle of black. In successive rings toward the circum- 
ference of the disk the relative amount of black is reduced, 
but the relation of black and white in each ring is constant 
throughout the width of the ring. If such a disk as this 

FIG. 31 

is rotated by means of a color mixer, there will be pro- 
duced about the center of the disk one black ring, and 
outside of this a series of gray rings differing from each 
other in intensity in such a way that each succeeding ring 
is somewhat lighter than that which lies nearer the cen- 
ter. The gray rings produced during rotation will not 
seem, as they are in reality, uniformly gray throughout 
their widths. That part of each ring which lies nearest 
to the inner, darker ring will seem relatively light, while 
that part of the ring which is nearer to the next outer ring 
will seem dark. These effects are due to contrast, for 
wherever a gray ring is in immediate contact with a light 


ring it will seem darker and, conversely, where it is in 
immediate contact with the darker ring it will seem lighter. 

If a second disk is prepared, as shown in Fig. 31, B, 
with fine lines drawn between the successive rings, the 
contrast effect will be very much weakened if not indeed 
entirely destroyed. The lines of demarcation between 
the successive rings prevent the eye from traveling freely 
across from one ring to the other and consequently dis- 
turb the contrast. 

Similar disks may be prepared to illustrate color contrast. 
Thus, as shown in Fig. 31, C, a disk may be prepared, 
the outer portion of which is blue, the inner portion of 
which is red. If now between these two large portions 
of the disk a middle ring be inserted which is made up of 
one-half of blue and one-half of red, there will be produced 
when rotation begins a central ring of purple. This cen- 
tral purple will seem very red on its outer circumference 
and very blue on its inner circumference, because at 
these two limiting circumferences contrast with the neigh- 
boring field is pronounced. 

Reference was made a number of times in connection 
with the last exercise to the fact that the eye does not 
respond to colors in the same way after it has been for a 
time in the dark. It is true in general that all color mix- 
tures and color contrasts are very much modified by a 
change in illumination. The general principle is that 
the reds and the closely related colors tend to disappear 
wherever there is a reduction in the illumination, before 
the blues and closely related colors. Consequently, if a 
mixture is made of red and blue and the general illumina- 
tion is reduced, the mixture will show a stronger blue 
quality than it showed in the stronger intensity of light. 

The third supplementary experiment requires that 
various cases of color experience shall be compared with 


each other. Such comparison can be made by matching 
each experience with a combination of standard disks 
placed in a constant illumination. Thus, if a red is to be 
compared with other reds, establish in each case an equa- 
tion by matching the first red with a combination of 
standard disks and expressing the combination of disks 
in a formula. Then find the formula for the second red 
and so on. 

The best method of demonstrating the difference be- 
tween color mixture of the type which has been dealt with 
in this exercise and the type of mixture which is involved in 
combining pigments, is to prepare a disk which has been 
covered with a wash of pigments in the following manner: 
Let the outer portion of the disk be given a coat of blue 
over half of its surface. Let the other half of the outer 
portion of the disk be covered with yellow. Let the inner 
portion of the disk be painted with a mixture of the blue 
and yellow pigments. The result of mixing these two pig- 
ments is a vivid green. If the disk is now placed upon a 
color mixer and rotated, the central green portion will not 
be modified in any way by the fact that the disk is ro- 
tated. The outer portion of the disk will, on the other 
hand, show the regular complementary relation between 
blue and yellow; that is, it will show as a result of the mix- 
ture a decided gray. 

After thus demonstrating the fundamental difference 
between pigment mixing and mixing by the method of 
successive stimulations, the nature of pigment mixing 
may be explained by the use of colored glasses, the results 
of such a test being exhibited in Fig. 32. Set up in a ray 
of sunlight a piece of blue glass. Examine with a prism, 
or, better, with a spectroscope, the light which passes 
through this blue glass. It will be found that the glass 
has deprived the white light of most of the red, orange, 



and yellow components which it originally contained. 
The blue, green, and violet components of the white light 
will be strong. It will thus he seen that the blue color 
of the glass is due to the fact that only a portion of the orig- 
inal white light is allowed to come through the glass. 
Similarly, it will be found on examining the light which 
comes through a yellow pane of glass that the orange, 
yellow, and green components of white light are strong. 
There is a great reduction in intensity of the reds, blues, 

Yellow Blue 

Fio. 32 

and violets by the yellow glass. If, now, light which has 
been allowed to pass through the yellow glass passes 
afterwards through the blue glass, it will be obvious that 
only that color which passes easily through both glasses 
will survive the double process of absorption through the 
two glasses. The rays of light which thus pass through 
both kinds of glass are predominantly green in quality. 
Whatever other rays escape absorption are so completely 
submerged by the strong green rays that the light seems to 


be altogether green. This result agrees with the experi- 
ment of mixing yellow and blue pigments described above. 
Indeed, the blue and yellow pigment particles which were 
used in making the color washes may very properly be 
regarded as a series of small transparent particles of blue 
and yellow media, which deprive the light which passes 
through them of all rays except the green rays in the same 
way that yellow and blue glass would. The effect of 
these pigment particles is due to the fact that the light 
which falls upon the surface of the color wash is carried 
downward into the pigment particles for a certain distance 
and is then refracted back again after having been partially 
deprived of its rays. Combinations of pigments other than 
blue and yellow can be worked out in analogous fashion. 
It is to be noted in general, however, that most other pig- 
ments are not as transparent as the blue and yellow pig- 
ments. There is, therefore, in general a great reduction 
in the intensity of light whenever pigments are mixed 
with each other. 



Exercise IV requires very little apparatus. The head- 
rest which was described on page 32 may again be 
utilized in this exercise, or the simpler devices suggested 
in the same exercise will serve. The eye shield is de- 
scribed on page 31. The measuring rods can be con- 
veniently held in clamps of the kind described in the 
general statement on page 9. In measuring distances 
on the rod it will be found convenient to have the ex- 
perimenter move some bright object, like a strip of white 
card, along the rod until it seems to the observer to be in 
the right position to mark the boundary of the object 
which is being measured. 

Drawing figures to scale is facilitated by the use of sheets 
of coordinate paper. This renders the reduction to any 
desired unit very easy. Thus, if the object is 24 cm. 
long and 12 cm. wide the successive measurements will 
be as follows: 

At 3 M .... 18 cm. long by 9 cm. wide 
At 2 M....12 cm. " " 6 cm. " 
At 1 M.... 6 cm. " " 3 cm. " 

If now the scale chosen for the drawing on the coordinate 
paper is 1mm. =2 cm., the length and width of the object 
can be represented by two lines drawn one under the other, 
the first being 12 mm. long, the second being 6 mm. long. 
Perpendicular lines should then be drawn from the centers 
of each of these lines to represent the distance from the 



object to the eye. This distance will be represented in 
the figure as 20 cm. The form of the resulting figure is 
given in the author's Psychology, General Introduction, 
page 155. 

The formula, which is easily derived from these results, 
is as follows: 

} ( to the corresponding dimen- 

Any dimension f . . ,, , . 1 . , . . . .? 

. v i > is in the same ratio < sion of its projection at a 
of the omect ( ) j- A 

) { nearer distance, 

( the distance of ^ ( distance of the 

as < the object from > is to the < projection from 
( the eye ) ( the eye. 

The diagram and calculation can be directly related to 
the image on the retina by calculating how far back of 
the optical center of the lens the rays must be projected 
in order to reach the retina. In any given case the dis- 
tance of the retina from the optical center can not be di- 
rectly determined, but a general average of many eyes 
which have been measured justifies the general assump- 
tion that the diameter of the eye is 23 mm. and that the 
distance of the optical center from the front of the cornea 
is 7 mm. The full formula for the size of a retinal image 
is accordingly: 

Dimen- corresponding distance of object 

sion of : dimension of :: from front of : 16 mm. 

object image cornea -f- 7 mm. 

An excellent exercise for a class is to require the compu- 
tation of the size of a house one hundred meters distant 
which seems just equal in height to one cm. held at a dis- 
tance of 10 cm. from the eye. Also give the size of the 
retinal image of such a house. How large would the ret- 


inal image be if one moved away to twice the original 
distance ? 

Turning to the second part of the exercise, a. well-de- 
fined after-image may be secured by cutting in a piece of 
black cardboard a cross with legs about ten centimeters 
in length and one centimeter in width. This cross should 
be held between the eye and the bright sky and the ob- 
server should look steadily at one point on the cross for 
a minute. If the cross can not be prepared, a gas jet or 
other bright object gives a very good image. 

For comparisons of the apparent sizes of projections of 
the after-image, it will be found convenient to use a draw- 
ing-board on which has been tacked a sheet of paper 
marked off by heavy black lines into square decimeters. 
The drawing-board can be easily moved about from one 
position to others. If a drawing-board is not at hand, a 
similar drawing may be made on heavy cardboard. 

The formula derived above will be found to hold for 
all of the distances to which the after-image is projected. 

It may be found necessary, in order to bring out the 
after-image clearly as it is projected to the various posi- 
tions, to wink the eyes rapidly. 

The Aubert's diaphragm required for the third part of 
che exercise consists of two pieces of metal or wood each 
one of which has a right angle notch cut in one end, as 
indicated in Fig. 33, 1, 2 and 3, 4. 1, 2 may now be 
superimposed upon 3, 4 to any desired extent, and the 
result will always be a symmetrical diamond-shaped 
figure 1, 2, 3, 4. This Aubert's diaphragm should, for 
the purpose of the experiment, be mounted on a box. 
The box should be taken into a dark room and a light 
should be placed inside of the box. No other source of 
light should be present. Between the light and the 
Aubert's diaphragm should be a thick plate of milk glass 



which will thoroughly diffuse the light. The diaphragm 
will control the size of this diffusely lighted glass, which 
is the only object in the room visible to the observer. 
In addition to the preparations thus far described, the 
box should be mounted on runners so that it can be drawn 
backward and forward in the third dimension directly 
in front of the observer. Furthermore, a system of levers, 
represented in Fig. 33, should be attached to the parts of 
the diaphragm so as to permit a rapid adjustment of the 
diaphragm from both sides. Lifting the long handle H 

FIG. 33 

produces a double effect. First, it throws the short arm 
L, which is rigidly fastened to H, outward away from the 
center of the diaphragm opening. The part of the dia- 
phragm 1, 2 is drawn outward with L, for L is fastened to 
1, 2 by means of a pin which fits into a slot in L. The 
second effect produced by lifting H is to rotate the small 
cog-wheel which is fastened at 0, the center of movement 
in H and L. This cog-wheel in turn sets in motion a 
second similar wheel which has fastened to it the short 
lever M, at P. M is equal in length to L and is fastened 
by a pin and slot to 3, 4. By this means M is moved in a, 


direction opposite to L whenever H is lifted, and 3, 4 
moves in -a direction opposite to 1, 2. When // is moved 
downward 1, 2 and 3, 4 are closed in on each other. The 
advantage of the whole system of levers is that the center 
of the opening always remains at the same point, and the 
effect, so far as the size of the retinal image produced by 
the area is concerned, is the same as the effect produced 
by moving the area in a direct line nearer to the eye or 
further away. 

The result of this part of the experiment will be that 
the observer will make many mistakes, confusing change 
in distance with change in opening. The size of a retinal 
image will thus be shown to be complicated with interpre- 
tations of distance to such an extent that size and distance 
must be recognized as closely related forms of interpre- 


The first supplementary experiment can be tried in a 
way to secure an accurate record, by taking a piece of 
cardboard about 10 cm. square and fixing on its under 
surface a pin. This pin should not be allowed to perforate 
the cardboard but should be fastened wholly on the under 
surface. A convenient method of accomplishing this is to 
pass the pin through a small piece of gummed paper. 
When the pin has been drawn through this paper as far 
as possible, leaving only the head on the gummed side of 
the paper, the whole may be fastened by the gum to the 
bottom of the pasteboard. The pin will thus be held 
firmly in position but will not be seen on the upper side 
of the pasteboard. The observer should now hold this 
pasteboard with the pin hanging down from its under 
side at the level of his open eye, the other eye being cov- 
ered or closed. Let him now, by means of a lead pencil, 


indicate on the upper surface of the cardboard the point 
at which the pin seems to him to be located. A number 
of trials can be easily recorded in this fashion and it will 
be found that the distance of the pin from the eye is not 
accurately observed. After this let him make the same 
trials with both eyes open. 

The reason why the threads required for the second 
supplementary experiment described in the Laboratory 
Manual, should be enclosed in a box is that it is important 
to exclude shadows and to cover up the fastening points 
of the threads. Any shadows which are allowed to fall 
upon the threads give an indication of the differences in 
position of the differently illuminated threads. Further- 
more, unless some device is employed to prevent the 
observer from seeing the points at which the threads are 
fastened, he will be able to judge of the different dis- 
tances of the threads from his eyes by means of the sur- 
faces which lie between the different points of attachment. 
Let threads, preferably of different sizes and of different 
colors, be drawn vertically through a box which is 40 or 
50 cm. cube. Let the upper and lower thirds of the open- 
ing of the box be covered with gray or black screens. 
The observer now looks through the middle third of the 
box and observes the threads which lie at different depths. 
He will be able to designate the different threads by their 
colors, and he will find, if he looks with only one eye, that 
it is almost impossible to give a correct judgment of their 
positions in the third dimension. Professor Jastrow has 
arranged a demonstration of analogous type by using 
cylinders of different degrees of curvature. A variety 
of blocks of wood were turned with different diameters, 
and these were placed in such a position that they were 
not covered with shadows as they would be in an ordi- 
nary environment. The light was thrown upon them 



from above. The result was that when they were 
viewed monocularly it was found to be very difficult to 
decide which one had the greatest degree of curvature. 

The next supplementary experiment will give the 
following result. When the pin-hole is held very near to 
the eye it is inside the limit for near accommodation of 
the lens. The result is that the light coming through this 
pin-hole will not be focused upon the retina of the eye. It 
will come to a theoretical focus behind the retinal surface 
(Fig. 34, F). Since it is not focused upon the retina there 
will be an area of diffuse illumination on the retinal sur- 
face (Fig. 34, RP). The lower part of this diffuse area will 

Fio. 34 

derive its light directly from the lower part of the pin- 
hole. If, now, a card is brought between the pin-hole and 
the eye and is gradually drawn upward from below so 
as to cut off part of the pin-hole, it will obviously cut 
off first those rays which come from the bottom of the 
pin-hole (Fig. 34, PP) and fall upon the lower part of 
the diffuse circle on the retina. It is a fact of general 
experience that all rays of light which affect the lower parts 
of a retinal image are under ordinary circumstances de- 
rived from the upper parts of objects. Put in other 
terms it may be said that the direction in which we refer 
all rays of light falling upon the retina is determined 


by a line drawn from any retinal point in question through 
the optical center of the lens (Fig. 34, PW). Under the 
circumstances of this experiment to interpret the lower 
part of the retinal image as related to the upper part of the 
object is misleading, for the simple reason that the object 
has not been brought to a focus upon the retina as are the 
objects seen under ordinary conditions. Thus, as seen 
in Fig. 34, the light PP from the lower part of the pin-hole, 
and the light RR from the upper part of the pin-hole, come 
to the theoretical focus F, crossing the retina at the points 
P and R. The image will be interpreted according to the 
ordinary methods of interpreting retinal images as if 
PP came from the direction VV and will consequently be 
regarded as coming from the upper part of the figure, the 
ray of light affecting R being interpreted as though it came 
along the axis XX. A very convenient method of making 
this demonstration is to prepare a short tube about 4 cm. 
long, one end of which is entirely closed except for a small 
pin-hole. Directly in line with this pin-hole a small pin 
should pass through the wall of the tube in such a way 
that it can be pushed across the line of light coming 
through the pin-hole. 

When the eye moves from a remote object to a point near 
at hand the observer will note that the apparent size of 
the remote object undergoes a change such that it seems 
to be somewhat smaller than when it was observed di- 
rectly. If the new center of fixation is the finger or a pen- 
cil held near the face, a further experiment may be tried 
by keeping the eye fixed upon the pencil or finger as it is 
brought nearer and nearer to the eye. During this move- 
ment of the center of fixation toward the eye, the apparent 
size of the remote object will continue to grow noticeably 
smaller. An explanation which has sometimes been 
offered of these facts is that the finger or pencil as it moves 


toward the eye gives a larger and larger retinal image, 
whereas the remote object continues to give a retinal image 
of the same size as at first. Since the finger or pencil is 
the more clearly seen object it furnishes a standard of 
estimation, and since it is interpreted as constant in size 
in spite of the changes in the retinal image, the more re- 
mote object is by contrast interpreted as growing smaller. 
The fatal objection to this explanation is that exactly 
the same changes in apparent size in the remote ob- 
ject appear even when there is no movement of the 
near object. Merely looking from a remote object to 
a near point of fixation gives the result that the remote 
object seems to grow smaller. Indeed, the pencil or finger 
is merely a device for securing fixation upon a point near 
at hand. If the observer is capable through voluntary 
effort of fixating the eye on a series of points near at hand, 
he can, without the aid of a pencil or finger, secure all of 
the results described above. The contrast between the 
near object and the remote object can therefore not be 
the explanation of the facts observed. A more adequate 
explanation can be given by assuming that the center of 
near fixation is the point at which all objects in the field 
of vision seem to be located; that is, there is in monocular 
vision no clear discrimination of differences in depth. 
Consequently, when the eye fixates a near point, remote 
objects seem to be drawn up to this near point. Even if 
there is some suggestion of difference in depth because of 
shadows or other secondary characteristics, yet the re- 
mote object seems to be nearer to the point of fixation 
than it is in reality. Since the retinal image from the 
remote object is constant in size and since it seems to fol- 
low the movement of the near point of fixation, it will be 
interpreted as growing smaller and smaller as the near 
point of fixation approaches the eye. This experiment 


is very similar to the experiment described above, which 
dealt with the projection of the after-image, and the prin- 
ciple of interpretation is just the same. 

The earliest experiments on monocular accommodation 
and its resulting sensations were tried by means of threads 
suspended in front of the observer. A uniform gray 
surface was spread out before the observer and he was al- 
lowed to look at this uniform surface through a shield 
which permitted only monocular vision. Across this 
monocular field was drawn a thread. The extremities of 
the thread passed out of the field of vision above and be- 
low. The thread was now moved backward and forward 
and the observer was required to state whenever he was 
able to distinguish its movement in depth and he was also 
required to state the direction of the movement. In some 
cases two threads were used, one being suddenly replaced 
by a second which was either nearer or further away than 
the first. This mode of experimentation obviated the 
necessity of pushing the single thread backward and for- 
ward, and it made possible more sudden changes in the 
position of the thread. 

The difficulty with this whole method of experimentation 
is that the thread in spite of its small diameter is sufficiently 
large to give an appreciable retinal image. When now 
the thread is moved backward and forward, its retinal 
image changes in size according to the general laws of 
visual perspective. This change in size constitutes a 
secondary criterion which may be used by the observer in 
judging its relative position. It became evident in the 
course of experimentation that a method should be de- 
vised by which some sort of object should be presented in 
the field of vision that could be moved backward and for- 
ward without changing the size of its retinal image. 
Such an object can be produced if the line to be fixated 


is the boundary line between two areas in the field of vision. 
Since the boundary line is a geometrical line and has no 
width it will not change in width as it is moved backward 
and forward. If this line is made long enough so that it 
always extends through the whole field of vision in the 
vertical, it will obviously not change in apparent length. 
The method of producing such a line is as follows: Let 
a uniformly illuminated field be set up in a dark room. 
The illumination of this field may be either from shielded 
lamps which throw their light only upon the gray surface 
leaving other objects in the room dark, or by means of a 
plate of milk glass which is illuminated from behind. At 
a suitable distance in front of this illuminated field the 
observer is placed behind a screen through which he can 
look with only one eye. Such a screen as this should 
usually be provided with a tube through which the ob- 
server must look. The tube serves better than a single 
opening in the shield to limit the field of monocular vision 
so as to exclude all of the objects in the room except a por- 
tion of the uniformly lighted field. Half way across 
this circular monocular field of vision is projected a black 
screen with a sharp edge. This black screen should not 
be lighted at all on the side which is turned toward the 
observer. Since it cuts off one part of the observer's 
field of vision it leaves only a part of the former circular 
field visible, and gives a sharply defined line between the 
black field and the gray semicircular surface. If now 
this black shield is moved backward and forward, the 
boundary line between the black and light portions of the 
monocular field will constitute an object of invariable 
dimensions. The ability of the observer to recognize 
the position of this boundary line can be measured exactly, 
as in earlier experiments his ability to recognize the posi- 
tion of the thread was measured, by determining the 


distance through which the line must be moved in order 
that its change in position shall be recognized. 

There is one criticism to be made of certain of the inter- 
pretations which have been given to the results of these 
experiments. It has sometimes been assumed that the 
covered eye is excluded as a source of sensation and that 
the experiment deals with purely monocular phenomena. 
The fact is that the closed eye exercises in many ways a 
large influence upon the open eye. Indeed, in certain in- 
dividuals it has been shown that the kind and degree of 
influence exercised by the closed eye introduces very com- 
plex factors into the total situation. Thus, it has been 
shown that there is in some cases a tendency for the 
closed eye to relax somewhat in its muscular tensions and 
to take a position somewhat divergent from that which 
it would assume if it were open and fixating the same ob- 
ject as the uncovered eye. On the other hand, while there 
is some tendency toward divergence in the covered eye, 
there is unquestionably a tendency for this eye to fixate 
the same point as does the uncovered eye. All these 
tendencies of movement contribute sensory factors and 
also complicate the behavior of the open eye, as has been 
amply shown by photographic records of the eye's move- 

A great deal of experimentation was undertaken in the 
early years of experimental psychology to determine the 
minimum visibile or least recognizable distance between 
two points. The simplest method of making this deter- 
mination was to draw two points or two parallel lines upon 
a sheet of paper and gradually move this paper further 
and further away from the observer. As the points were 
moved further and further away the retinal image of the 
intervening space gradually decreased in size until finally 
the two images seemed to flow together. The size of the 


image could now be computed by the formula given on 
page 65. By this method it was shown very clearly that 
the least perceptible distance between points or lines 
when seen in direct vision is very much less than the least 
perceptible distance between points or lines which are 
seen indirectly. The just perceptible distance is in- 
fluenced in a measure by the amount of illumination. 
If the experiment is made by drawing black lines on a 
white sheet of paper and the white sheet of paper is rela- 
tively little illuminated, the lines will flow together much 
sooner than they will when the paper is brilliantly illu- 
minated ; that is, the intervening white space in such a case 
will be recognized much longer when it has a high light 
intensity than when it is very faint. It is probably true 
that if an intermediate area were illuminated with suffi- 
ciently great brilliancy it would never disappear between 
two bounding dark surfaces. Standard illumination is 
therefore quite as essential to the success of this experi- 
ment as any other condition. 

The value of such experiments as these is very largely 
that they determine the acuity of the organ of sense. The 
importance of the results is that they throw light on the 
general problem of the differentiation of sensory surfaces. 
Since the tests are of value chiefly in determining the con- 
dition of the organ of sense they have been worked out in 
a much more practical form by those who are constantly 
called upon to use them for diagnostic purposes. The 
oculists have devised a number of methods for testing 
the acuity of vision. Charts made up of letters of differ- 
ent sizes are commonly utilized to test vision. These 
charts are placed in a given illumination and the ob- 
server is tested with reference to his ability to read the 
different sized letters. The greater the acuity of vision 
the smaller the letters which he will be able to recognize. 


These visual charts have been modified sometimes in 
order to adapt them to use with illiterate persons. In- 
stead of letters, lines drawn in such relation as to consti- 
tute the three sides of a rectangle are placed in various 
positions. Thus the open side of the rectangle is some- 
times placed above, sometimes below, sometimes at the 
right, and sometimes at the left. The observer is required 
to indicate which side of the figure is open. These figures 
are printed in various sizes upon the chart, and the acuity 
of vision is tested by the observer's ability to recognize 
correctly the figures at standard average distances. 



The student after observing double images, as directed 
in the Laboratory Manual, should represent the relations 
involved by means of a diagram. 

Fig. 35 represents the relation between the two eyes 
and the two objects N and R, which are located at different 
distances in depth from the observer. Rays of light from 
N passing through the optical centers of the lenses of the 

Fio. 35 

two eyes pass to the points F and F f , which represent the 
two foveas in the two eyes. The fact that the images 
from N fall on the foveas indicates that the two eyes are 
fixating this near point. The rays of light from the re- 
mote point R will also pass through the optical centers 
of the lenses and will fall upon the two retinas at the 
points R and R'. It will be noted that R and R f are both 
on the nasal sides of the retinas. Any point other than 



the point of fixation, if at the same depth, casts its image 
on the retinas of the two eyes in such positions that in 
one eye the image will fall on the temporal side of the fovea 
and in the other eye on the nasal side of the fovea. The 
condition which is represented in Fig. 35 is therefore one 
in which fusion of the two images can not follow as it 
would under ordinary conditions of observation of plane 
surfaces. The two retinal images R and R', falling both 
on the nasal sides of the two retinas, do not fuse. The 
observer will consequently see these two images as separate. 
Some observers have difficulty in recognizing such double 
images. This is due to the fact that objects not in the 
center of the field of vision are vague because they are not 
clearly focused, and they are consequently for the most 
part neglected. Furthermore, there are a great many 
observers who habitually neglect to a greater or less de- 
gree one of their retinal images. If the double images 
R and R' are recognized as separate by the observer, it will 
be further noted that the image for the left eye appears to 
lie on the left side of 'the point of fixation. This can be 
clearly demonstrated by closing or covering the left eye. 
Correspondingly, the image for the right eye will seem to 
lie on the right side of the center of fixation. Under 
these conditions the double images are called uncrossed 
double images. 

If the same diagram is used to represent the reversed 
condition where the eyes are fixated upon the more re- 
mote object R, the points R and R' must be treated as the 
two foveas. F and F f will therefore represent two points 
on the temporal sides of the retinas. These images will 
be seen as double and will seem to lie on sides of the 
center of fixation exactly opposite from the eyes to which 
they belong; that is, the image for the right eye will seem 
to lie on the left side of the center of fixation and the image 


for the left eye will seem to lie on the right. These double 
images are known as crossed double images. 

The clearest explanation of the apparent position of 
these double images is that the objects not at the center of 
fixation are attracted toward the center of fixation, as in- 
dicated in an earlier exercise, page 72. The points 
where the lines of light from the objects cross the plane 
of the point of fixation will accordingly determine the ap- 
parent position of the double images in space. In Fig. 
35 the planes are represented by two lines 00' and SS'. 
If the center of fixation is the near point, the lines of light 
from the more remote object cross the nearer plane at 
the points and 0'. If the center of fixation is the more 
remote object, the lines of light from the near points must 
be projected, as indicated in the dotted lines, to the re- 
mote plane, which they will intersect at the points S and 
<S'. Double images are essentially monocular phenom- 
ena. They appear in the binocular field but they are due 
to an absence of fusion of the images. They can be 
studied with reference to their characteristics, and it will 
be found in general that they do not give data necessary 
for localization in depth. 

A very good introduction to the second part of Exercise 
V is to require the student to indicate by means of a dia- 
gram the form and position of the images on the retinas 
of the two eyes from a solid object. (See Psychology, 
Gen. Intr., page 159, Fig. 48.) The student should also 
draw a similar figure indicating the character of the retinal 
images derived from a hollow object. 

Any solid object will serve to show the difference be- 
tween the images of the right and left eyes. It is desirable, 
however, that the object observed be small enough and 
simple enough so that the student can draw it for the later 
parts of the experiments, hence the suggestion of a trim- 



cated pyramid. Small blocks of wood should be sawn 
out with a square base of 5 cm. on each side. The pyra- 
mid should be from 4 to 6 cm. in height and the small 
truncated surface at the top of these models should be 
2 cm. on each side. The figure will be much more easily 
used for the purposes of making drawings if the surfaces 
of the block are painted black and the edges are marked 
with a fine white line. The observer should now set this 
model on a piece of paper and should close one eye and 
examine the model with the single open eye. He should 
indicate by points on the paper, first the dimensions of 

FIG. 36 

the base, and then the apparent position of the edges of 
the small upper surface when this is projected to the plane 
of the paper. Without changing the position of his head, 
let him now open the eye which a moment before was 
closed and close the eye which was open. Again, let him 
plot the dimensions of his figure on the paper, noting 
especially the change in the apparent position of the pro- 
jection of the upper square. He should now draw the 
figure in full with a result similar to that shown in Fig. 
36, where A is the image for the left eye, B the image for 
the right eye. 
The next step of the experiment consists in preparing 



the apparatus for the fusion of these two figures. The 
apparatus in its simplest form is a mirror stereoscope. 
Let two boards be set up as indicated in Fig. 37, 1 and 2. 
These boards are held most firmly in position by means 
of a common base BB. On the base BB should be set up 
two mirrors placed at angles of 45 with reference to 
1 and 2, and at an angle of 90 with respect to each other. 
Rays of light from board 1 will now be reflected forward 
from the mirror 3, while rays of light from board 2 will be 
reflected by the mirror 4. If the two eyes of the observer 

FIG. 37 

are placed so as to look into mirrors 3 and 4, the observer 
will see the rays of light, which in reality come from the 
boards 1 and 2, as if they came from some point in space 
behind the mirrors. If, now, instead of plain boards 
1 and 2, the figures drawn from the truncated pyramid 
are placed in proper position on these boards, the observer 
will see the images as if they came from a single object be- 
hind the mirrors and will fuse them into a single figure in 
space. Fig. 38 shows the fusion in diagrammatic outline. 



A stereoscope of this form can be set up without any 
special construction of new parts if the laboratory is sup- 
plied with a sufficient variety and number of clamps and 

The mirror stereoscope suffers from certain defects 
which interfere somewhat with its use. In the first place, 
any ordinary mirror has three reflections from its surfaces. 
One reflection comes from the upper surface of the glass, 
a second from the lower surface, and a third from the silver 
surface. The latter two images are so near each other that 

they do not interfere noticeably with the experiment. 
The image from the outer surface of the glass, however, 
will disturb the ordinary observer very decidedly. There 
is no easy method of obviating this difficulty. A method 
which may be adopted consists in securing highly polished 
metal surfaces and using these for mirrors. Such metallic 
mirrors are, however, likely to tarnish, and it is difficult 
to keep them ready for use. A better expedient, if one 
is conveniently near a mirror-manufacturing establish- 
ment, is to secure small mirrors before the silver surface 


used on the back of the glass has been covered with paint, 
as it is in the finished mirror. The silver surface which is 
intended for use through the glass may be used in such 
unfinished mirrors on the side turned away from the glass, 
provided the mirror is carefully handled. From this 
unfinished surface only one reflection will be received. 
Such a mirror tarnishes very soon and can not be re- 
newed without destroying the silver surface altogether. 
A new mirror, must, therefore, be provided for each series 
of experiments. A second limitation of the mirror stereo- 
scope consists in the fact that it neglects a natural habit 
of adjustment which appears in all observers. In general 
it is true that the lenses in the two eyes are habitually ad- 
justed in focus to the point of fixation; that is, if the two 
eyes are converged upon a point one meter distant from 
the eyes, the lenses will be so accommodated as to focus 
light which comes from the distance of one meter. This 
type of adjustment very often does not suit the conditions 
presented by the mirror stereoscope. Thus, if the center 
upon which the eyes are converged in looking at the stereo- 
scopic object lies at a certain distance back of the two mir- 
rors, the lenses may very frequently be called upon to 
focus light which comes from a distance very different from 
that of the point of convergence, because the light which 
enters the eye originates from the figures tacked to the 
boards of the stereoscope rather than from the point of fix- 
ation. The only condition under which the natural ad- 
justments are the proper adjustments, are those under 
which the center upon which the two eyes are converged 
is at exactly the same distance from the eyes as the figures 
from which the images on the two retinas are derived. In 
order that this particular case may be attained, it is de- 
sirable that the simple form of stereoscope described above 
be elaborated so as to make the distance of the images 


from the two eyes variable. This can be done either by 
making the boards 1 and 2 (Fig. 37) adjustable in their 
distance from the mirrors or by making the angle between 
the two mirrors adjustable. An adjustment of the two 
boards 1 and 2 can be provided for by mounting these 
boards on wooden blocks, which in turn fit into runners 
placed along the base of the stereoscope, BB. 

The mirrors 3 and 4 may also be mounted upon separate 
standards so as to be adjustable in their angle with refer- 
ence to each other. In an even more elaborate form of 
apparatus, the base BB is divided into halves and so 
mounted that each half can be rotated about a single point 
between the two mirrors. In this way the angle of the 
mirrors with reference to each other can be changed 
without disturbing the relative angular positions of draw- 
ings 1 and 2 with reference to their respective mirrors. 
Experiments may in this way be made on movements of 
convergence without disturbing the other relations in- 
volved, especially those between the figures tacked to 1 
and 2 and the mirrors which reflect their images. 

The mirror stereoscope may be made in almost any 
desired dimensions. A very convenient size is one in 
which the base BB is 1 meter long and 25 cm. wide. The 
mirrors in such an apparatus should be 5 cm. square. 
Smaller stereoscopes may be made to give excellent results. 
The figures must, however, in such cases be made much 
smaller, and the student can not deal as easily with models 
in preparing the figures to be fused. 

If the two drawings represented in Fig. 36 are fused in 
the stereoscope, the result will be that the observer will see 
a single solid truncated pyramid all sides of which are sym- 
metrical. If, now, the wide side of one of the figures is 
covered, the fused figure of the pyramid will change some- 
what in its character. It will no longer seem to be a sym- 


metrical figure, nor will the side which is now seen monocu- 
larly be solid as in the original fused figure. There will 
be a continuous outline of the pyramid on the side for 
which one figure is covered, but this side will seem flat, 
and since it is the wide side of one of the drawings which 
was covered the figure seen will seem narrow. Con- 
versely, if the narrow side of the figure is covered, the cor- 
responding part of the pyramid will seem flat and wide. 
The fact that the monocularly seen parts of the pyramid 
are respectively too narrow and too wide shows that the 
binocularly seen figure has a width which is a compromise 
between the width of the two monocular figures. When 
both the wide and narrow sides are present, they give 
through fusion a compromise width, and they also yield 
the characteristic of solidity which was not present in either 
of the monocular elements which entered into the total 
figure. These observations give the clearest evidence that 
the fusion process is one in which all of the sensory factors 
receive due recognition, the resultant characteristics of the 
binocular figure depending upon compromises and fusions 
of the monocular elements.. 

Two figures with no lines in common can easily be pre- 
pared for the last part of this exercise by drawing on one 
card a series of vertical lines and on the other a series of 
horizontal lines. When the effort is made to fuse these 
two groups of lines the observer will see first one set of 
lines and then the other. In some cases the lines will not 
appear and disappear as a single group, but there will 
be a small area at which the vertical lines will appear 
in the midst of a horizontal field, and this area of 
vertical lines will gradually spread until the horizontals 
are for the most part submerged. Other figures can be 
prepared by using lines which extend in various direc- 
tions. Surfaces of different colors may also be useq to 



show the lack of complete fusion between entirely different 
kinds of fields. 


The stroboscope is an apparatus by means of which 
there is exposed to the eye of the observer a rapid succession 
of figures. These figures represent the successive stages 
of some activity, such for example as the flight of a bird or 
the movements of an animal or human being in walking 
or running. The essential 
condition for successful 
fusion of such a series of 
figures is that the eye shall 
see one image for an in- 
stant and shall then be 
supplied with an entirely 
different image, the first 
being covered so as to 
avoid any blurring or fu- 
sion of the two successive 
images. The stroboscope FIG. 39 

is usually made in the form 
of a cylindrical case in 
which a succession of vertical slits are cut. (For a 
simple hand stroboscope see Fig. 39.) Back of each 
of these slits is introduced a single figure. When now 
the eye looks through one of these slits it sees the single 
figure which lies behind it. As the cylinder is rotated 
this image is cut off after being seen by the observer, and 
as a new slit comes before the eye a second image is ex- 
posed to view, and so on. 

A recent observer has described a very interesting ex- 
periment, in which by means of the stroboscope the suc- 

From the catalogue of M. Kohl, 
Chemnitz, Germany 


cessive views of a solid object, which would be seen by 
looking at the object with a single eye first on the right side, 
then from directly in front and finally from the left side, 
are fused when viewed successively in the stroboscope in 
such a way as to give the appearance of solidity. There 
can be no doubt that a person who has only one eye de- 
rives from his head-movements a series of images from 
solid objects which he uses for the recognition of solidity 
in a way very similar to that in which the normal indi- 
vidual uses binocular differences. 

An elaborate apparatus which is in principle the same as 
the stroboscope is familiar to all who have seen the mov- 
ing pictures or kinetoscope pictures which are commonly 
used to reproduce series of movements. The kinetoscope 
uses a series of photographs which correspond to the differ- 
ent stages of the movement to be represented and pro- 
jects these photographs in rapid succession upon the same 
point on the screen. 

When, as suggested in the second supplementary ex- 
periment, the angle between the two mirrors is changed, 
the center of convergence for the two eyes will move in- 
ward or outward inward if the angle of the mirrors with 
respect to each other is decreased, and outward whenever 
the angle is increased. The angles through which the 
mirrors may be rotated are small. If these angles are 
made too large, fusion of the two images ceases and double 
images appear. With every change in the degree of 
convergence there will come a distinct perception of change 
in position of the object. There will also result a percep- 
tion of change in size such that the object will seem to 
grow smaller if the angle of the mirrors is decreased and 
the point of fixation is made to approach the observer, 
while the object will seem to -grow larger if the angle be- 
tween the mirrors increases and the point of fixation re- 


cedes. If the changes here under discussion are extreme, 
double images will result, as indicated above. 

Certain of the effects produced by changing the angles 
of the mirrors can be paralleled by simple experiments 
with natural objects when these are viewed in adjustable 
mirrors. Let two mirrors be held in exactly the same plane 
as shown in the full drawn lines A, B, in Fig. 40, and let 
some object be observed at i as it is reflected in these 

FIG. 40 

two mirrors. The apparent distance of the object from 
the observer will be equal to the total distance through 
which the light has traveled from the object to the eye; 
that is, it will seem to be as far behind the mirrors as the 
object is in reality in front of the mirrors. If, now, the 
two mirrors are so arranged, as indicated at A and B in 
Fig. 40, that the reflected rays which enter the right eye 
come from the right-hand mirror, while the reflected 



rays entering the left eye come from the left-hand mirror, 
the angle of convergence can be modified without modi- 
fying the actual distance through which the light travels. 
This can be done by changing slightly the angle between 
the two mirrors, as indicated by the broken lines L and 
D in Fig. 40. Since the lines of convergence are changed 
by this inclination of the mirrors, the point of fixation 
will also seem to change, coming to the point O 2 , and there 

Fir,. 41 

will result an apparent modification in the position and 
size of the object. The apparent change in position will 
follow the rule that whenever the angle between the two 
mirrors on the side of the observer is greater than 180, 
the object will seem to approach and at the same time to 
grow smaller. Whenever the angle on the side of the ob- 
server is less than 180, the object will seem to recede and 
grow larger. Only very small changes in the angle can 


be made without producing double images. The angle 
shown in the figure is much exaggerated. 

The third supplementary experiment requires the use 
of the pseudoscope. The simplest form of pseudoscope is 
one which was devised by Prof. George M. Stratton. It 
consists of a box, as represented in Fig. 41, which is sup- 
plied with three openings a,tA,B, and L. The distance be- 
tween A and B and between B and L should be about the 
ordinary distance between the two eyes of an observer; 
namely, between 5 and 7 cm. In order to make the box 
useful for a great number of observers, the holes at A, B, 
and L should be made in the form of ellipses about 1 cm. 
in the transverse axis and 2 cm. in the axis lying in the 
line of the three holes. In front of the hole B should be 
mounted a mirror, as indicated at M , at an angle of 45 
from the back of the box. A second mirror should be 
mounted at M' in the corner of the box parallel to the 
mirror M. If now the two eyes of the observer are brought 
into position before the openings A and B, obviously the 
left eye looking through A will see any object at the 
point D directly in front of the opening. The right eye, 
looking through J5, will not receive light directly from the 
object , but will receive its light only after it has been re- 
flected from the mirror M' into the mirror M and then 
through the opening B. (For convenience in drawing, the 
rays of light from the object and D to the two eyes are 
represented as parallel, that is, as coming from a very re- 
mote object.) The effect of this double reflection is to throw 
into the right eye the image that would be received by the 
eye placed in the position of the mirror marked M'. The 
right eye, therefore, gets its image from a position at the 
left of the left eye and the relation between the two eyes is 
thus reversed; for the right eye gets its image from what 
is relatively the left-hand side of the object, and the left 


eye gets its image from what is relatively the right-hand 
side of the object. The result of such an interchange of 
the images in the two eyes will be that solid objects will 
appear hollow and more remote objects will appear near 
at hand. It will be very difficult for the observer trained 
in the ordinary observation of solid objects to get the in- 
terpretation of the hollowness from familiar objects. It 
will be easier to get the interpretation of changes in rela- 
tive position, near objects appearing far away and remote 
objects near. The observation can be facilitated by using 
objects which are very little shaded, and by asking the ob- 
server to wink his eyes during the observation. This 
winking of the eyes makes the adjustment between the 
two eyes relatively easier. 

The pseudoscope described thus far may be transformed 
into a telestereoscope by using the two openings B and L 
instead of the openings A and B. The right eye, which in 
this case is looking through the opening B, will get its 
image not from directly in front, but from the position 
M'. The distance between the two eyes will then be ex- 
aggerated by the double reflection through the mirrors, 
and the object will be seen under a greater parallax than 
when the two eyes are observing it in their ordinary posi- 

There are a variety of forms of stereoscopes and pseudo- 
scopes ; perhaps the best known of the former is the lens stere- 
oscope popularly known as a toy. The lens stereoscope is 
somewhat more complex in principle than the mirror stereo- 
scope and for this reason is not as good for laboratory 
work. It can be used, however, and figures on a reduced 
scale can be prepared as required in the exercise. The 
lenses in such a stereoscope serve two distinct functions. 
From their shape and position they act first as prisms 
and deflect the rays. Secondly, in their capacity as lenses 



they assist the eye to focus the light from the figures. In 
order to make clear these two functions, Fig. 42 shows in 
full drawn lines two prisms in the positions occupied in 
the ordinary lens stereoscope by the lenses. It will be 

noticed that the two faces of the prisms turned towards 
the two eyes are in the same plane. The surfaces of the 
prism more remote from the eyes are oblique. Rays 
of light from A and 5, which represent the figures, will be 
deflected in passing through the prisms, as indicated in 


XXF and YYF'. The two eyes receiving the rays XX F 
and YYF' will be converged, and the observer will interpret 
the two rays as if they came from the single point R be- 
hind the true figures. If the figures A and B are of suit- 
able character they will fuse and the observer will see a 
single solid object at R. Obviously a conflict similar to 
that described in the case of the mirror stereoscope will 
arise between the reflex tendency of the lens of the eyes to 
focus upon R and the requirement that they focus the light 
from A and B. In this stereoscope 
the conflict is not overcome by the 
effort of the observer, but by plac- 
ing lenses in the position of the 
prisms. The lenses are placed in 
the position indicated by dotted lines 
in Fig. 42. Here they serve all the 
purposes of deflecting prisms and, 
furthermore, aid the lenses of the 
eyes in focusing the light from A 
and B, which light comes from a po- 
sition nearer than R, upon which the 
eyes are converged and for which 
the lenses are naturally focused. 

A great variety of figures suitable 
for use with the lens stereoscope can 
be purchased of any one of the 
stereoscope supply companies. Underwood & Under- 
wood of New York City, 19th Street and Fifth Avenue, 
have a collection of sterograms which cover most of the 
important phenomena of binocular vision. 

Another form of stereoscope is illustrated in Fig. 43. 
With some training an observer may accustom himself 
to the fusion of figures of the kind usually used in stereo- 
scopes without the aid of any apparatus whatsoever. If, 



for example, the image appropriate to the right eye is 
drawn on a card and the image for the left eye is drawn on 
a second card, and these two cards are placed in the posi- 
tions A and B, Fig. 44, A, and the two eyes are voluntarily 
crossed so that the axes of vision are along the lines AF 
and BF f , there will be seen at a single solid object 
which results from the fusion of the two images derived 
from A and B. If the two cards A and B are appropriate 
in form and nearer than the two eyes, fusion may take 
place as indicated in Fig. 44, B. The fused image will 

FIG. 44 

in this case appear at R. Such fusion requires practice 
in order to dissociate convergence and accommodation, 
for obviously in both cases the lenses of the eyes must focus 
light coming from a point other than the point of fixation. 
The whole matter of retinal rivalry is a fruitful field for 
investigation. There are undoubtedly great individual 
differences in observers in the ability to fuse different 
colors, and the strain which is produced by the effort to 
fixate two objects undoubtedly influences the rate of rivalry. 
Dr. Breese in his paper " On Inhibition," Monograph Sup- 


plement of the Psychological Review, No. 11, reports a 
number of experiments on retinal rivalry in which he has 
shown that the movements of the eyes are important ele- 
ments in determining the degree of fusion of two rival 
fields. There is one special case of partial fusion which 
properly belongs under the general head of retinal rivalry. 
If two fields which are otherwise suitable for fusion are 
different from each other in that one is black and the other 
white, there will be a unique type of fusion resulting in 
metallic luster. This metallic luster corresponds di- 
rectly to what is seen in ordinary experience under similar 
conditions. If one looks, for example, at a metallic sur- 
face which is illuminated from some definite source of 
light, this surface will reflect more light into one eye than 
into the other. This is equivalent to saying that one 
surface is dark while the other is relatively light. The 
observer who looks with two eyes at such a surface will 
not only see the surface but will also see the luster, which 
is shown by the stereoscopic experiment to be due to the 
inequality in the illuminations of the surfaces. If now 
the coloring of the surfaces differs not only in intensity 
but also in quality, the fusion becomes still more incom- 
plete with the results noted in rivalry. 

Methods of recording the rate of rivalry will be described 
in connection with Exercise XXII. These methods in- 
volve certain recording apparatus which will be in constant 
use in the second part of the course, but is not needed in 
general in this part. 

The exact point upon which the two eyes are converged 
when two images are fused by means of a stereoscope is 
impossible to define from direct introspection. It can be 
determined by finding out the positions of the lines of re- 
gard. The method of procedure, if one is using a mirror 
stereoscope, is to find on the two mirrors the points from 


which the rays enter the two eyes from what seems to be 
in the fused image a single point. This can be done by 
moving a lead pencil successively across the two mirrors 
until its point seems to coincide with the point chosen. 
The distance between the two points on the two mirrors 
should now be measured; the distance between the two 
eyes should be measured, and the distance between the 
eyes and the mirrors. If the distance between the two 
eyes is regarded as the base of a triangle at whose ver- 
tex the point of fixation lies, the distance of this point 
of fixation can readily be plotted by recognizing that the 
distance between the two points on the mirrors is the base 
of a second similar triangle which also has as its vertex the 
point of fixation. The distance between the two bases 
of the two similar triangles is the known distance between 
the eyes and the mirrors. The whole matter can be plotted 
as shown in Fig. 38 (page 83). When the lens stereo- 
scope is used instead of the mirror stereoscope, it is some- 
what more difficult to determine the point of fixation, be- 
cause the lenses are so near to the eyes. In this case some 
object should be interposed between the lens of the stereo- 
scope and the eye. The two points on the surfaces of the 
lenses near the eyes can now be ascertained which cor- 
respond to a single fused part of the object. Further 
procedure is similar to that followed in the case of the 
mirror stereoscope. 



The head must be held in a fixed position in this experi- 
ment. To this end the head-rest described on page 32 
may be used, or rods may be clamped to the chair in which 
the observer sits. In either case the space back of the 
head and around the ears must be left free. 

Any means of producing an easily controlled sound of 
small intensity will serve the purpose of this exercise, 
though telephones are so much better than any other form 
of sounder that they are explicitly referred to in the text. 
There is a child's toy known as a snapper, or sometimes 
called a telegraph sounder, which is very convenient. This 
toy can be made to produce a sharp single click by pressing 
upon a spring. The noise produced by this snapper is 
especially advantageous for this experiment, because it is 
simple in quality and is very little modified by the position 
of the snapper in front of the ear or behind it. 

The greater convenience of telephones as sources of the 
sounds consists in the fact that telephones can be easily 
controlled in their positions and the sound can be pro- 
duced in two or more telephones at exactly the same in- 
stant. In the parts of the experiment which require 
more than one sound at the same time, the use of two 
snappers is difficult, because the experimenter must de- 
pend upon his own ability to produce the sounds at ex- 
actly the same instant. Two or more telephones can be 
utilized by passing the same electric current through them. 
When this electric current is made or broken both tele- 



phones will act in exact harmony. The watch case tele- 
phones, so called, are better than the larger receivers for 
the purposes of this experiment, They are smaller and 
can therefore be better held in position. As many of 
these telephones as are necessary for the experiment 
should be put in circuit with one of the batteries referred 
to in the general introduction as yielding about one am- 
pere of current. In the circuit should also be placed a 
mercury switch; this can be used in making and breaking the 
circuit and does not interfere with the experiment by itself 
producing a click. 
The simple method 
of using the make- 
and-break current 
to sound the tele- 
phones has one dis- 
advantage. Unless 
the telephones are 
tuned exactly to the 
same pitch, it is 
sometimes possible 
to distinguish them 
because of their 
difference in quality. This difficulty can be overcome by 
putting into the circuit a simple interrupter. When the 
interrupted current is allowed to pass through the tele- 
phones, the result will be a series of vibrations of the 
telephone diaphragm rather than a single click. Minute 
differences in the pitch of the telephone diaphragm will 
then be entirely overcome, and the sound will be somewhat 
clearer and easier to locate than with the single clicks. 
The principle of the interrupter is one which is used 
in a great many pieces of apparatus employed in the labor- 
atory, and it may be described at this point in some de- 

Fio. 45 
From the catalogue of C. H. Stoelling Co., Chicago 


tail. The rod A, Fig. 45, is held firmly in position by the 
heavy post shown at the left of the figure. The rod car- 
ries at its right end a long platinum needle which extends 
downward into a mercury cup. From the mercury cup a 
wire extends to the electromagnet which is held above 
the rod A. The electromagnet is insulated from the rest 
of the apparatus and is connected at the end of its coil 
opposite to that at which it is connected with the mer- 
cury cup, with the battery. The other pole of the battery 
is connected with the rod A through the metallic post. As 
soon as the platinum needle is pushed downward so as 
to dip into the mercury cup a complete electric circuit is 
made which passes through the magnet. As soon as the 
magnet becomes active it pulls the rod A upward and 
thus tends to draw the platinum needle out of the mercury 
cup and to break the circuit. As soon as this is done 
the magnet is no longer active, and the rod A by its own 
elasticity oscillates back again so as to make a contact 
between the mercury cup and the platinum needle. This 
renews the electric current, causes the magnet to pull the 
bar A up again, breaks the circuit, and the whole opera- 
tion is again repeated. The rate at which the electric 
current will be interrupted depends upon the rate at 
which the rod A oscillates. In some other forms of inter- 
rupters the contact here provided by the mercury cup 
and platinum needle is made by means of an elastic 
platinum wire and a fixed metallic plate. The use of 
platinum in all of these contrivances is necessary because 
there is an electric spark each time the current is broken, 
and this spark would very soon burn any other metal. 
Furthermore, in the mercury contact it is desirable that 
the mercury surface be covered by a drop of alcohol or 
water which will take up the fumes from the burning 


Contrivances for holding the telephones in position 
may be made as elaborate as desired. The most conven- 
ient method of fastening the telephones and measuring 
their positions is to prepare an auditory cage. In its 
simplest form this cage may be made of wooden hoops. 
A somewhat more elaborate form of auditory cage is rep- 
resented in Fig. 46. The simpler form made of wooden 
hoops is constructed on the same principle and will be 

FIG. 46 

readily understood from the following description of the 
more elaborate form. A circular metallic ring AAA is 
supported by similar metallic semicircles CC and DD. 
These semicircles cross each other and are fastened firmly 
at a point M just over the middle of the circle AAA. 
From this point M a bar is carried upward, as shown at 
EE. This can be held at any point that may be convenient. 
In the figure, EE passes through the tube F which is 
held in the bracket BB. EE is held in position by the 


collar G which is clamped to it by means of a set screw 
H . If the collar G is loosened the shaft EE may be ad- 
justed so as to bring the cage to a higher or lower point to 
suit a given observer. The whole cage may be rotated 
within the carrier F, so that any point on the circle AAA 
may be brought into any desired horizontal position. A 
circular disk R is clamped on the outside of the carrier 
F by means of a screw. This disk has graduations marked 
upon it to indicate degrees, and a pointer is brought up 
from the shaft EE, as indicated at L, so that the position 
of the pointer and therefore of the whole cage can be read 
on the circular disk R. 

The telephones are now fastened on the frame AAA t 
or on the side pieces C or D, and the cage 
can be rotated as the experimenter desires. 
A convenient method of fastening the 
telephones to the frame is represented in 
Fig. 47. C represents a section of the 
frame. The bent metal plate A hooks over 
the top of the frame. Below there is a 
screw /S, which controls an adjustable tooth 
H. When H is screwed against A, the whole is firmly 
clamped to C. When, on the other hand, S is loosened, 
H falls down and permits A to be carried to any part of 
the frame desired. A is riveted to the back of a telephone 
which is by this means easily clamped to any part of the 

A simpler form of holder can be devised by using a piece 
of spring brass which can be bent into a clip and slipped 
over the frame. 

Another attachment necessary for the first part of the 
experiment is a rod which can be fastened to the cage in a 
horizontal position. Along this rod slides a carriage to 
which a telephone may be clamped. 


The various parts of the frame of the cage should be 
graduated in degrees and the rod should be graduated in 

The whole cage can be supported in various ways. It 
is desirable to support it at a distance from any large 
surface which could serve to give an echo of the sounds. 
The best device for supporting any cage is a high stand 
made of a heavy base and a piece of iron pipe, such as 
is used for steam or water connections. For ordinary 
purposes of class work, it is possible to fasten the cage 
to a bracket extending from a wall of the room. The 
experiment will be somewhat interfered with by the echo 
from the wall, but if the sounds are not too intense the dis- 
advantages of this method will be reduced to a minimum. 

Any form of cage which offers large reflecting surfaces 
of wood or metal is objectionable; the metal strips used for 
the cage should, therefore, be as narrow as practicable. 
The fact that the experimenter must stand near the cage 
is also an objection to the device thus far described. 
This difficulty can be overcome by more elaborate devices 
for adjusting the cage by means of strings carried to a dis- 
tance from the cage. 

A very elaborate auditory cage is described by Professor 
Seashore in the Psychological Review, Vol. X, pp. 64-68, 
and in Monograph Supplement of the Psychological Re- 
view, No. 28, pp. 1-5. 


The following series of observations was reported for 
the first part of the experiment, when one telephone was 
kept in a fixed position 50 cm. from the left ear in the 
same horizontal plane as the ear and in a line pass- 
ing through the two ears, and the other telephone 


was moved in the same line through various distances 
from the right ear. 

Distance of 

second tele- 

phone from 

Apparent position 

Apparent distance from head 

right ear in 



Right, about 5 to rear 

Near head, 10 cm. distant 


Right, about 20 to rear 

Further away, perhaps 



Right, about 50 to rear 

Still further, 30-40 cm. 


Rear, about 10 to right 

Still further, 40-50 cm. 


Rear, perhaps 5 to right 

Still further, 50 cm. 


Rear, about 10 to left 

50 cm. distant 


Rear, about 25 to left 

Same as last 


Left, about 45 to rear 

Same as last 


Left, about 30 to rear 

Same as last 


Left, about 20 to rear 

Same as last 

In the second part of the exercise the distances through 
which the telephones must be moved were found by vari- 
ous observers as follows : 

Starting in median plane in 
front or behind and moving 
to right or left 

Starting opposite one ear and 
moving forward or back- 










Avg. 25 
Avg. 40 

4.3 * 


If complex tones are used it will be found that there is 
much more complete discrimination of positions in front 


of the head and behind. This is probably due to the fact 
that the pinna in reflecting the sound into the ears modi- 
fies the sound somewhat by reinforcing certain of its 
components. This action of the pinna as a resonator is 
different according to the direction from which the sound 
comes. , 

The second supplementary experiment which uses tun- 
ing-forks in direct contact with the head is of some interest 
because the bones of the skull transmit the sounds directly 
to the two inner ears. The apparent localization of the 
fused resultant is often inside of the skull. 

Oblique positions give complications of the results re- 
ported for horizontal positions. Variations in the qual- 
ity of components have been reported as giving variations 
in the vertical or oblique positions of the fused resultant. 
Variations in the pitch can be produced by using dia- 
phragms of different thicknesses in the telephones, or by 
loading an ordinary diaphragm with lead. If the differ- 
ence between the two sounds is too great they will not fuse. 

If reflectors are fastened to the ears the effect of the 
pinnas as resonators is entirely changed, and as a result 
the apparent location of the sound will be modified. 

Improvement of auditory localization through practice 
is reported by Pierce in his "Studies in Auditory and 
Visual Perception." 

The use of three telephones complicates the problem 
by introducing one factor which may dominate the whole 
experience (if, for example, one source of sound is out- 
side of the median plane and the other two are in that 
plane). This is a case which differs radically from color 
and light fusions. 



The simplest method of producing tones for the pur- 
poses of this exercise is by means of a set of Quincke 
tubes. A Quincke tube is represented in Fig. 48. It 
consists essentially of a blower B held by means of a wire 
in front of a sounder *S. The sounder is stopped at one 
extremity by means of a cork. If the blower is held in the 
right position at the mouth of the tube, a fairly pure note 
may be produced by blowing through B. 
The air for the blower can be supplied 
most readily by the mouth. If it is de- 
sired to blow more than one tube at a 
time, as in the experiment, a mouthpiece 
can be made from tin with a single open- 
ing at one end to be inserted in the mouth, 
and two or more openings of appropriate 
size to fit over the blowers of the de- 
sired number of tubes at the other end. 
Quincke's tubes can be made to produce 
FlG 48 tones of relatively high pitch. 

A second simple means of producing musical tones of 
desired pitch is to employ two or more chromatic pitch- 
pipes. The form of pipe diagrammatically represented 
in Fig. 49 serves the purpose very well. The reed pipe 
PP is supplied with a sliding damper DD. The foot of 
the damper rests on the reed RR and may be drawn up 
or down so as to make the reed longer or shorter, and the 
note of the pipe consequently lower or higher. If a scale 



is attached to the damper at some point, any desired 
note may be produced and read on the scale. Two such 
pipes with a combination mouthpiece, such as that de- 

FIG. 49 

scribed in connection with the discussion of the Quincke 
tubes, will serve very well the purposes of the exercise. 

Even with the simple devices thus far described, it is 
desirable to provide an air supply which shall be more 
constant in intensity than that which can be produced by 

FIQ. 50 

blowing with the mouth. A simple type of air reservoir 
consists of a large vessel indicated in Fig. 50. This outer 
vessel A should be two-thirds filled with water. A second 


vessel of like form should be inverted in the position B y 
and tubes D and E should be introduced through the outer 
vessel and led upward into B above the level of the water. 
Air may be introduced into the vessel B through the tube 
D from any convenient source. A hand or foot bellows, 
a foot pump, or a mechanical pump, may be used to sup- 
ply air for this purpose. When the air is needed to blow 
the pipes it should be drawn from the second tube E. 
The reservoir B will act as a storage reservoir and 
through its weight it will also tend to force the air out 
under a constant pressure. The pressure can be increased 
by placing on top of the reservoir various weights. 

If it is desired to use a current of air for any lengthy 
experiments, automatic mechanical contrivances can be 
introduced into the supply pipe Z), so as to open or 
close the supply pipe and start the pump according to the 
height of the reservoir B. A stop-cock at H should open 
when the reservoir B approaches its lower limit and a con- 
nection from K should start the pump. The air will then 
be allowed to enter and raise the tank B. The stop-cock 
should be gradually closed during this raising of the tank 
B, until finally the reservoir is completely filled ready for 
the experiment, at which time the pump may be stopped. 

Titchener and Whipple describe a double reservoir 
(American Journal of Psychology, 1903, pp. 107-1 1 2). Two 
tanks corresponding to B in the figure are chained to- 
gether so that one rises as the other falls. The falling 
tank is provided with weights. By transferring the 
weights from one movable tank to the other, a stream of 
air under very uniform pressure can be supplied with 
no interruption except that involved in transferring the 

A mechanical pump can be used with a storage reser- 
voir without the pressure tank here described. In that 


case, since the air is delivered from the storage reservoir 
at an irregular pressure, a double regulator valve should 
be secured which will so regulate the outflow that the air 
can be delivered at the point of use at a uniform low 
pressure. Such valves are to be had through any mechanics' 
supply house. They are manufactured by Wm. Boekel 
& Co., 518 Vine St., Philadelphia. (Catalogue E.) 

Even when blown steadily by means of an automatic 
air-pump, Quincke's tubes and chromatic pitch-pipes are 

Fio. 51 

less constant in their qualities than organ pipes. A sys- 
tem of organ pipes mounted on a keyboard will be found 
to be a very satisfactory means of supplying tones. The 
practical difficulty in securing this kind of equipment 
consists usually in getting the keyboard for sounding the 
pipes rather than in securing sets of pipes. Fig. 51 shows 
the mechanism of a keyboard, such as is commonly used 
in pipe organs. The air chest AA has two openings, one 
very small one at C, and a large one at K. The one at C 
is always open and tends to keep the air in the small 


outer chamber B with which it communicates at the same 
pressure as the air inside the air chest A A. The open- 
ing K is closed by the valve KJ. This valve consists of 
an arm J with a fulcrum at its center. At the end K is a 
felt foot fitting over the opening which leads out of the air 
chest. At the opposite end the arm J is fastened to a piece 
of flexible sheepskin M. The valve K J, is held in position 

Fio. 52 
From the catalogue of Kohl, Chemnitz, Germany 

by the pressure of air within the air chest and by the small 
supplementary spring L. The valve K J is opened indi- 
rectly by reducing the pressure on the outside of the sheep- 
skin M. This is accomplished by opening the valve D 
which communicates with the outer air. The valve at D 
can be opened by pressing on the finger piece E. As soon 
as the pressure in B is reduced, the sheepskin M is forced 


outward and opens the valve K. This allows the air to 
pass out of the air chest AA into the organ pipe P. 

Elaborate instruments for producing series of tones are 
manufactured. Appunn's "tone messer" is a reed pipe 
in which the successive pipes differ from each other, not 
by the conventional intervals, but by two vibrations in 
each case. 

Stern has worked out an apparatus called a "tone-vari- 
ator" for producing continuous series of variations in 
tonal quality. This is represented in Fig. 52. The me- 
tallic cylinder C has a piston Kb which is raised and low- 
ered from below by a cam. The volume of air within 

FIG. 53 
From the catalogue of C. H. Stoelting Co., Chicago 

the cylinder is thus varied by any desired amount, and 
the change can be read directly on the scale connected 
with the cam. A pipe, A, which extends obliquely 
down to the mouth of the cylinder, excites the 
cylinder whenever an air current passes through it. This 
apparatus has the advantage of producing not merely a 
variety of tones, but every possible modification of pitch. 
The range of a single cylinder is limited and a series of 
cylinders must be provided for any extended experiments. 
Instead of wind instruments such as have been described 
up to this point, string instruments may be used. The 
most available instrument of this sort is the sonometer 
represented in Fig. 53. Over a resonator box are stretched 


wires. Under these wires are adjustable bridges which 
may be placed in various positions along scales marked 
on the resonator. The wires are kept under tension 
either by weights, as shown in the figure, or by means of 
screws. The tones from the wires are clear and of long 
duration. They differ in timbre according to the point 
at which the wires are plucked or bowed when excited. 

Another means of producing continuous series of tones 
is by means of weighted tuning-forks such as are shown 
in Fig. 54. Tuning-forks give very much more reliable 
tones than any other type of apparatus, and the tones are 
simpler in quality than any that can be secured either 

FIG. 54 

After Wundt. " Grundauge der physiologischen Psychologic," 
5th Ed., VoLII, p. 82 

from reeds, organ pipes, or string instruments. On the 
other hand, the tone from a tuning-fork is by no means as 
intense as the tone from a wind instrument. The intensity 
of the tone can be increased by mounting the forks on 
resonating boxes. 

The simplicity of the tones from tuning-forks is a matter 
of importance for the complete success of the experiment 
with the fused tones. If the tones are at all complex, 
the degree of fusion will depend quite as much upon the 
other components as upon the chief or fundamental tone. 
For this reason the use of organ pipes or stringed instru- 
ments is likely to result in a discrimination of tones even 


when they would fuse completely if they were simple. 
The production of pure tones requires the very best 
tuning-forks. It is relatively easy to produce a tuning- 
fork which shall give a fairly pure tone, but if the mate- 
rial in the tuning-fork is irregular in its temper or in its 
density, the fork will vibrate in different ways in its dif- 
ferent parts and there will result a complex tone. For 
this reason the best quality of tuning-forks will be found 
indispensable for reliable scientific work with tones. 
If such tuning-forks are secured, the second problem 

FIG. 55 

From Wundt's " Grundziige der physiologischen Psychologic," 
5th Ed., Vol. II, p. 82 

is to provide for the transmission of the sound from these 
forks to the hearer. It is not a satisfactory method to ex- 
cite the two forks successively and then attempt to judge 
of their fusion, for the successive sounding of the forks 
will give the observer opportunity to discriminate the 
sounds more completely than he could if the sounds 
had been presented to him together. The arrangement 
shown in Fig. 55 makes it possible to excite the forks 
without giving the observer an opportunity to hear them 
until they are both in vigorous action. A rubber tube 


(S Z 0) should be carried through a heavy wall from one 
room into the next. Indeed, it is preferable that the 
tube be carried through a still greater space. In any case 
it should be of rubber rather than of rigid material. At 
the end O, should be seated the observer with the tube 
placed against the ear. For this purpose a glass tube 
may be carried from the rubber to the ear of the observer, 
this glass tube being shaped so as to fit the ear of the ob- 
server. At the other end of the rubber tube there may 
be, in addition to what is shown in the figure, a stopcock 
by means of which the tube can be completely closed while 
the experiment is in preparation. This tube should be 
connected with two or more resonators, such as are shown 
at I, II, and III. A very convenient method of procedure 
is to have two resonators, one of which is fixed in size, 
the other of which has a piston so arranged that it can be 
adjusted within the resonator, reducing it to the proper 
size for any one of the forks of a given octave. In front of 
the resonators should be clamped the forks which are to 
be used in the experiment. When all is in readiness for 
the experiment the observer should be warned that he is 
to hear the sound in a few seconds. The tuning-forks 
should be excited either by striking them or by bowing 
them with a violin bow. The shutters in front of the reson- 
ators should be drawn aside; the rubber tube should be 
opened so as to transmit the sound to the observer. It 
may be found necessary, if the wall between the observer 
and the apparatus is not thick, to completely disconnect 
the rubber tube from the resonators during the preparation 
for the experiment. Furthermore, in order to prevent the 
sound from being transmitted by the solid walls, it is 
sometimes necessary to suspend the resonators and the 
tuning-forks from springs so that the sound may be pro- 
duced upon an isolated bridge. 


The intensity of the tones can be varied by exciting the 
source less vigorously or by cutting off part of the conduct- 
ing channel to the observer. If the source is a Quincke 
tube or reed-pipe the tube which carries the air into the 
tube can be partly closed. The effort to cut down the in- 
tensity of a sound after it has been produced is always 
complicated by the fact that the walls of the conductor 
are as likely to carry the vibrations as the columns of air 
inside of the conductor. It is better, therefore, to regu- 
late the sound at its source. 

For the third part of the experiment two chromatic 
pitch-pipes or two Stern's variators are the best devices. 

In working with intervals, the simplest procedure is to 
have one source adjustable (for example, either a chro- 
matic pitch-pipe or a tone-variator) and the other fixed. 
Now sound the interval between the two, and immediately 
after vary one of the sources slightly and repeat. It will 
be found that the ability to recognize intervals is in gen- 
eral very highly developed. 


A table showing the distribution of judgments in the 
first experiment of the exercise with three untrained ob- 
servers is as follows: 



one tone 

one tone 


two tones 

one tone 



















































































The degrees of fusion here indicated do not agree in de- 
tail with Stumpf's table, but they give a sufficiently clear 
indication of the difference between CC' and CG on the 
one hand and such combinations as CB and CD on the 
other hand. Stumpf recognizes five degrees of fusion: 
from the most complete in the octave, CC', through 2d) 
the fifth, CG, 3d) the fourth, CF, 4th) the pure thirds 
and sixths, CE and CA, to 5th) minor sevenths and 
other combinations, CB and CD. 

When one tone is stronger than the other, fusion is more 
complete, the stronger tone dominating the whole percept. 

The ability to discriminate tones differs greatly with 
different observers, ranging from one vibration to half a 
tone. If will be found in some cases that the observer 
can discriminate tones before he can tell which is higher. 

The recognition of intervals, as pointed out above, is 
often better than the recognition of tonal differences. 


The apparatus for developing combination tones is the 
same as that described. Quincke tubes serve very well for 
difference-tones, adjustable forks for beats. For pur- 
poses of demonstrating difference-tones the double whistles 
which are supplied as bicycle warning whistles sometimes 
give a very marked difference-tone. 

The last supplementary experiment which deals with 
highest and lowest tones requires apparatus not described 
thus far. 

Series of steel cylinders of different lengths are manu- 
factured which make it possible to secure, by simply strik- 
ing the different cylinders, very high metallic tones which 
are pure in quality because all of their overtones are much 
beyond the range of ordinary hearing. 


The apparatus most convenient for securing high tones 
is Galton's whistle, the most elaborate form of which is 
shown in Fig. 56. This whistle consists of a pair of 
small tubes, D and E. The air is forced into D and so 
against the tip of E from the 
rubber bulb. Vibration is set 
up in E. The length of the air 
column in E is regulated by 
means of a piston controlled by 
the graduated screw G and F. 
The rate of the vibration will be 
determined by the length of the 
air column in E. In order to 
prevent transverse vibrations, the 
opening between D and E is 
regulated by the graduated screw 
B. There is a simple form of 
this whistle which is subject to 
certain errors. The more elab- 
orate form supplied by Edelmann 
obviates these errors and gives a very accurate measure- 
ment of the highest tones. 

For the lowest audible tones, simple rods may be used. 
These may be held firmly in a vise and shortened or 
lengthened until the limits of audibility are reached. 
A simple rod made for this purpose is supplied by Appunn. 

FIG. 56 

From the catalogue of Edel- 
mann, Munich, Germany 



Very little apparatus is required for Exercise VIII. 
A convenient method of marking the arm at the outset of 
the experiment is to provide a rubber stamp, or, better, an 
electrotype stamp, which has on its face a figure marked 
off into squares two millimeters on each side. Such a 
stamp as this can be used with the ordinary stamping 
pad and a very good impression can be produced on the 
skin. An impression from the same stamp should be 
repeated in the notebook of the experimenter as a means 
of making a record of the differences in 
sensitivity discovered during the explora- 
tion of the skin. Fig. 57 shows the im- 
pression from such a stamp. 

Points for the exploration of the skin 
can be provided in the simplest fashion 
by beveling the end of a short brass 
rod to a blunt point. The point should 
not be too fine, as it will be likely to prick the skin and 
cause discomfort during the experiment. On the other 
hand, it should not be so blunt as to spread the stimulation 
diffusely over the skin so as to affect a series of points at 
the same time. 

In order that the hand of the experimenter may not 
come into direct contact with the metal rod and thus serve 
as a means of rapidly cooling or warming it, the rod should 
be passed through a cork and the experimenter should 
handle the rod by means of this cork. Such a metallic 


Fig. 57 


point as this may be placed in ice water if it is desired to 
experiment with cold spots, and in water of various other 
degrees of temperature if it is desired to work with warm 

More convenient forms of temperature points can be 
made as indicated in Fig. 58. The handle furnishes a 
means of holding the point which may be renewed as often 
as desired from a supply which is kept in cold or warm 

Instead of the solid points, cylinders may be used. 
These can be filled with water of any desired temperature 
and will hold their temperature longer than the rods. In 
the most elaborate forms of temperature cylinders, a cur- 
rent of water is drawn from a remote reservoir, where it is 
kept at the desired temperature, 

and passed continuously through *^ -V *) 

the cylinder. 

For work with pressure spots, FlG 58 

bristles of various sizes may be 

fastened to wooden handles by means of sealing-wax, or 
they may be held in screw clutches which fit into the 
handle shown in Fig. 58. A clutch for this purpose 
is made by splitting a screw and putting the bristles be- 
tween the two parts of the screw. The screw should be 
made in the form of a cone. When a nut is screwed up 
on this taper, the sides of the screw will be drawn together 
and will firmly grasp the bristle. 


Experiments dealing with the relativity of tempera- 
tures can be readily prepared by using dishes of warm 
and cold water. The observation is so closely related to 
the facts of common experience that no detailed discussion 
of the matter is necessary. 


For the second supplementary experiment, prepare a 
series of light corks which are equal to each other in 
weight but differ in size, and ascertain by tests which one 
will give the greatest apparent pressure. 

Experiments on the tongue can be made by drying the 
surface thoroughly and afterward applying drops of the 
substance with which the organ of taste is to be stimulated 
by means of a camers-hair brush. 

One test for sensitivity of the skin to which reference 
is not made in the Laboratory Manual is the test for sensi- 
tivity to pain. The apparatus for such a test is known as 
an algometer. It is virtually an inverted spring balance, 
and consists in a hollow wooden cylinder which has a 
coiled spring inside. Against this coiled spring there is a 
piston which extends out beyond the end of the wooden 
cylinder. The end of the piston is held against the part 
of the skin which is to be tested, and the wooden cylinder 
which serves as a handle is now pressed downward toward 
the skin. The pressure on the cylinder acts upon the piston 
through the spring and tends to push the piston against 
the skin with as much force as is exerted by the coiled 
spring within. If a scale is marked on the wooden cylinder 
and a pointer is attached to the piston, this pointer will tend 
to travel up the scale as the pressure is exerted on the 
piston, showing how much the spring is brought into play. 
The amount of pressure necessary to produce pain at any 
given point can thus be read off directly from the scale 
as it would be read on a spring balance. 



For the first part of this exercise a simple tracing may 
be made by laying the hand on a piece of paper and out- 
lining its form with a pencil. The outline should be sup- 
plied with lines to indicate the positions of the knuckles, 
and certain of the prominent features of the hand and 
arm. If it can be had, a still better map with which to 
work is a life-sized photograph of the arm. A more 
elaborate method yet is to secure a plaster of paris cast 
of the arm and hand. The latter device is necessary only 
when a long series of experiments is to be tried. A pho- 
tograph as a substitute for the rough outline is advan- 
tageous even for simple demonstration experiments. 

For producing the sensations in the first part of the ex- 
ercise a simple wooden point can be used. The points 
can be marked by ink made of aniline dye and water. 

As a means of supplying the stimulation at two points 
on the skin an ordinary drawing compass may be em- 
ployed. It is not absolutely necessary, as prescribed in 
the text, that the points should be made of hard rubber, 
but if they are not so made there is large possibility that 
the observer will be distracted from time to time through 
the excessive stimulation of temperature spots on the skin. 
Even with a compass, additional points of rubber or bone 
or wood may easily be fastened over the metallic points, 
and thus the difficulty which might otherwise arise through 
excessive stimulation of the temperature spots may be 
eliminated. If drawing compasses are employed, it will 



be necessary for the experimenter to measure the distance 
between the points by comparison with a scale. 

It is obviously more convenient to have the measure 
directly connected with the points, so that when an adjust- 
ment is made a direct reading of the distance of the points 
from each other can be seen at a glance. Fig. 59 represents 
a simple apparatus very similar to the one that was em- 
ployed by Weber in his experiments. The long bar A is 
graduated into millimeters. One point is directly con- 
nected with the extremity of the graduated bar. Over 
the bar there travels an adjustable arm which can be 
placed at any desired distance from the fixed point. The 

Fio. 69 

distance between the two points can now be read directly 
on the scale. Such an apparatus is known as an aesthesi- 

Certain refinements can be introduced into an apparatus 
of this sort. Thus, instead of using hard rubber points 
Von Frey has suggested that bristles be used which 
shall be of equal thickness, so that the two points 
can be pressed against the skin with equal intensity. 
Such an addition as this to the sesthesiometer makes im- 
possible any inequality of pressure at the two points. 
Others have suggested that the sesthesiometer points be 
held in some kind of sliding handle so that they will 
always rest against the skin with the weight of the appar- 
atus. These refinements, however, are unnecessary. 


With a little practice the experimenter can learn to set the 
two points down with equal pressure and with the great- 
est precision in the matter of the time of contact of the two 

In the course of a series of experiments with the two 
points it is customary to introduce what has been called 
a confusion experiment. This consists in stimulating 
the observer with a single point. It is sometimes found 
that an observer gives a judgment two points when, as a 
matter of fact, he is being stimulated only with a single 
point. This is doubtless due in many cases to the fact that 
other points on the skin are more or less irritated by in- 
ternal conditions or by after-images of early tactual stimu- 
lations. Indeed, after one has been experimenting for 
some time with a given region of the skin it is almost im- 
possible to decide whether a given experience is the re- 
sult of a present stimulation or an after-image. These con- 
fusion results will be found to be sources of constant error 
especially with untrained observers. It is generally well, 
in case these errors become too numerous, to begin with a 
distance between the two points so large that the observer 
has no difficulty whatsoever in recognizing it, and then to 
gradually reduce the distance until the two points are no 
longer clearly recognized. The degree of certainty in the 
observer's judgments of the separateness of the two points 
may also be considered in determining the threshold for 
two points. One fact which comes out very clearly from 
these results is that the skin threshold is not a definite sen- 
sory matter, but is rather the result of a complex perceptual 

If in addition to the judgment of separateness the ob- 
server is required to judge the direction of two points or a 
line upon the skin, it will be found that this judgment of 
direction is vague in the extreme, especially in untrained 


parts of the skin. More investigation of this matter can 
readily be made by the methods here described. 

In the part of this exercise where it is required to use 
lines instead of points, the simplest means is to prepare a 
series of cardboard lines. These lines should be cut at 
lengths varying from a single millimeter to any desired 
length. More convenient than cardboard lines will be 
found a system of thin lines made of hard rubber, varying 

FIG. 60 
From an article by Henri, in the "Archives de Physiologic," 1893, pp. 619-627 

in length, as do the cardboard lines, from a single milli- 
meter to lengths of from 5 to 6 centimenters. 


Results of experiments in localization when single 
points are touched are shown in Fig. 60 and also in the 
article by Pillsbury in the American Journal of Psychology, 
1895, pp. 42-56. 


The results in millimeters of the sesthesiometer experi- 
ment on three observers were as follows: 

Observer A 

Observer B 

Observer C 

















Finsrer. . 

When a continuous line is used, it will be found that it 
is distinguished from a point when it is less than half as 
long as the distance which must lie between two points 
in order that they may be distinguished as two. 


Distances across the arm are always recognized earlier 
than distances in longitudinal directions. 

The supplementary experiments on different parts of 
the body and on the effects of practice require no new 
apparatus or methods of procedure. 

If a point is moved across the skin it will be found that 
the fact of movement can usually be recognized before 
the point has moved over the distance required for the dis- 
crimination of two sesthesiometer points. The fact that 
perceptual discrimination of different stimulations can be 
more readily made during the movement of the point has 
been dwelt upon by certain writers as clear evidence that 
the perception of motion is a separate process from that 
of general spatial discrimination. Elaborate devices have 
been constructed for moving points across the surface of 
the skin at a uniform rate. These devices have usually 
been supplied with means of varying the rate from very 
rapid to very slow. They consist in principle of a moving 
carriage that carries a pointer which rests with its full 


weight upon the skin. The weight of such a point can be 
regulated by adding to the natural weight of the pointer 
or by counterbalancing it so as to make it lighter than it 
would be if it rested with its full weight upon the skin. 
For the purposes of ordinary experimentation much pro- 
ductive observation can be made by using a wooden point 
and moving it with the hand, not attempting to employ 
the more elaborate forms of apparatus. 

Many experiments with the semicircular canals have 
been tried upon animals. The canals have been de- 
stroyed in various ways and the animal's behavior after 
the disturbance of its canals has been carefully observed. 
Certain pathological human cases also serve for similar 
experiments. Experiments with normal human beings 
can be tried by directly stimulating the canals. If an 
electrode is applied to the outer surface of the skull di- 
rectly back of the pinna, effects may sometimes be pro- 
duced of slight dizziness or in some cases movements of 
the head may be induced which are analogous to the cor- 
rective movements on the part of animals when abnormal 
conditions of stimulation are produced through vivisec- 
tion in the canals. Such experiments by stimulation are, 
however, not agreeable and are hardly suitable for general 
experimental courses. A better method of demonstration 
and experiment for general classes consists in observing 
the effect upon the general recognition of bodily position 
when the excitation in the canals is unusual. Thus, let 
an observer stand with closed eyes, and point directly in 
front of himself. Now let him take three or four steps to a 
neighboring wall or blackboard where he is required to 
indicate the point which seems to him to be directly in 
front of the original position from which he started. This 
experiment should be repeated several times in order to 
determine the variations which will appear under these 


undisturbed conditions. After the average error of point- 
ing out a position which is directly in front has been 
ascertained, let the experiment be repeated by the same 
observer with the head turned sharply in some unusual 
direction. Thus, let the observer hold the head with the 
left ear as near to the left shoulder as possible or, con- 
versely, with the right ear near the right shoulder, or let 
the head be thrown back as far as possible, or let the neck 
be bent forward as far as possible. It will be found in 
these cases that the points selected on the blackboard 
after several steps forward differ decidedly from the 
points indicated in the earlier experiments. There may 
be involved in these experiments factors other than the 
sensations from the semicircular canals. The strain upon 
the muscles of the neck and the effect of a changed posi- 
tion of the whole head are undoubtedly to be considered 
in explaining the unusual direction of movement. 

Measurements of the ability to recognize changes in 
the position of the body can also be made by means of a 
tipping and rotating table. The observer lies on such a 
table and is gradually tipped so that the head rises and the 
feet are lowered, and record is made of the point where he 
first recognizes the change in position from the starting- 
point. Again the table is rotated so that the observer's 
head moves clockwise or counter-clockwise, and the angle 
is measured through which he must move in order to recog- 
nize the fact of movement and the direction. Such meas- 
urements as these probably deal with changes in the pres- 
sure of the lymphatic fluids in the semicircular canals. 
They may also include certain skin sensations. Much 
will depend upon the rate at which the movement is made. 



The simplest means of securing weights for this experi- 
ment is to take cartridge shells and ordinary shot and load 
the shells so that they are as heavy as desired. Other 
forms of cylinders can be prepared consisting of hard 
rubber cylinders or metallic cylinders. The cylindrical 
form is distinctly advantageous for all of the work because 
cylinders can always be picked up in the same way by the 
observer. Hard rubber is better than metal as the latter 
introduces temperature sensations. 

These weighted cylinders can be most readily presented 
to the observer by means of a small rotating table. The 
arm of the observer is supported above this rotating table 
within easy reach of the weights placed upon it. The ex- 
perimenter operates the table, bringing first one weight 
and then the other under the observer's hand. If no such 
rotating table as this is provided, the experimenter may 
place the arm of the observer in a convenient position and 
set the weights successively under the hand. 

Methods of treating the results of such tests are given in 
the text of the Laboratory Manual. 

Faint sounds can be most simply produced by means 
of a small body which is allowed to fall against a plate 
of glass or metal. A convenient body for this purpose is a 
small pith-ball. A suitable scale may be fastened at the 
side of a glass plate, and the pith-ball held in a funnel or 
taken in a pair of forceps which can be brought to any 
desired height on the scale. When the funnel or forceps are 



opened, the pith-ball will fall through a known distance 
to the plate of glass or metal below. The sound which 
it produces will be sufficiently faint to allow a considerable 
range of variation either in the height of the fall or in the 
distance of the object from the ear, and a measurement 
can thus be secured of the intensity of sound just neces- 
sary for auditory recognition. If the apparatus is moved 
further and further away from the ear in order to produce 
a fainter sound, it is not necessary to have a scale connected 

FIG. 61 
From the catalogue of Zimmermann, Leipzig, Germany 

with the glass plate. The forceps or funnel may he held 
in a fixed position at a certain height above the glass. 
The distance through which the pith-ball falls will thus be 
in every case the same, the variation in intensity of sound 
being produced by the distance of the apparatus from the 
observer's ear. Fig. 61 shows an elaborate device of this 

A second form of audiometer consists in a telephone 
which is connected with the secondary coil of an induction 


coil. The secondary coil is made adjustable so that it can 
be varied in its position with reference to the primary coil. 
The current which is to pass through the primary coil is 
controlled by means of a make and break key. If now 
the secondary coil is placed at a given distance from the 
primary, and the current in the primary is made and broken, 
a sound will be produced in the telephone. If the sound 
is of sufficient intensity for the observer to recognize it, 
the intensity of the current in the secondary coil should be 
reduced by moving this coil further away from the primary, 

Fig. 62 

and the experiment should be repeated until the distance 
is found at which the observer is unable to recognize the 
faint sound produced in the telephone. The distance 
between the coils constitutes a measure of the auditory 

Physicians use a much simpler test than either of these 
in determining for diagnostic purposes the sensitivity of 
the ears. They use the tick of a watch or a faint whisper, 
and measure the threshold in terms of the distance to which 
the sound must be removed from the ear in order that it 
may become too faint to be recognized. 


A simple form of photometer is shown in Fig. 62. Two 
sources of light are set up on the blocks A and B. Con- 
venient sources of light for this purpose are candles. Be- 
tween the two sources of light is a shield D. There is set 
up at C a rod which will cast shadows on the screen as 
shown at S and S'. The blocks A and B should be moved 
by the experimenter until the shadows S and S' seem to 
the observer just noticeably different in intensity. The 
difference in the distance of the two blocks from the rod 
can be determined, and this distance constitutes a measure 
of the just perceptible difference in illumination necessary 
for the observer's recognition. If now the total illumina- 
tion of the room is changed, it will be found that the dis- 
tance of the blocks from the rod necessary to produce a 
just perceptible difference in the shadows has also changed. 
In order to control easily the total illumination of the 
room as required in this experiment, the apparatus 
should be set up in a room that can be darkened. 

Other photometers are constructed in such a way that 
light from different sources falls upon two neighboring 
plates of milk glass. Such plates of milk glass should be 
separated from each other by an opaque metal strip. 
They should be placed so that they can both be seen by 
the observer at the same time. The sources of light should 
now be so adjusted that the illumination of one plate of 
glass can be recognized as just noticeably greater or less 
than the illumination of the other plate. The principle 
here employed consists in a direct examination of the illu- 
minated surface, rather than in the comparison of shad- 
ows as in the photometer shown in Fig. 62. In the second 
form of photometer, as well as in the shadow photo- 
meter, the measurement is made in terms of the distance of 
the sources of light. These measurements can be reduced 
to an absolute physical basis by recognizing the general 


physical formula that the intensity of lights is inversely 
proportional to the square of their distances. 

Other forms of photometers are numerous and can be 
arranged without difficulty. The principle of all such in- 
struments is sufficiently illustrated in the one or the other 
of the forms described. 


In the first supplementary experiment the comparison 
of differences is suggested. Such a comparison can be 

worked out with auditory sen- 
sations by arranging four balls 
so that they can be dropped 
from different heights. A cer- 
tain difference in height be- 
tween balls 1 and 2 will produce 
a known difference in the in- 
tensity of their sounds. Balls 3 
and 4 are also dropped from 
known heights, the difference 
between their heights being ad- 
justed in the course of succes- 
sive trials to satisfy the listening 
observer. The difficulty in car- 
rying out this experiment arises 
from the great difficulty of pro- 
viding balls that are just alike 
and uniform plates on which the 
balls may fall. Furthermore, 
the balls must be dropped by some sort of mechanical 
device which shall prevent them from rotating and shall 
be noiseless. Such an elaborate apparatus has been 
worked out in the I^eipzig Laboratory and is described in 

FIG. 63 

From Wundt's "Grundziige der 

physiologischen Psychologie," 

5th Ed., Vol. I, p. 513 


full in the Philosophische Studien, 1892, Vol. VII. The 
essential part of the apparatus, namely that by means 
of which the ball is dropped without rotation, is shown 
in Fig. 63. 

It is much simpler to determine the threshold for just 
perceptible difference in sounds. Fig. 64 represents a 
pendulum apparatus designed for this experiment. Two 
pendulums with ivory bulbs at their ends are so pivoted 
that they can be lifted to suitable distances in front of 
scales and then allowed to descend against a block. As 
the pendulum rebounds it is caught either by the hand or 

FIG. 64 
From the catalogue of Diedrich, Goettingen, Germany 

by a felt catch. The second pendulum is dropped in 
like manner through a slightly different distance, and 
the second sound thus produced is to be compared by the 
observer with that produced by the first fall. If the 
difference is not perceived, the experiment should proceed, 
the difference between the two distances through which 
the pendulums fall being slightly increased. 

An apparatus known as an olfactometer is used for ex- 
periments with intensities of odors. Let a glass tube TT 
be supplied with a nasal bulb as represented in Fig. 65 at 
N. This tube is held by a handle. Over the tube T is 
slipped a second larger tube MM, which is lined on the 


inside with a layer of paraffin. The tube T is represented 
in the figure with its outer tube MM in such a position 
that the air drawn into the nose during inspiration will 
have been exposed for a time to the paraffin surface of 
MM and will have taken up any odor which MM tends 
to give out. If MM is drawn further along T, a greater 
surface will be exposed and the odor will be relatively 
more intense. The amount of surface exposed in MM 
is accordingly a measure of the intensity of the odor en- 
tering the nose. 

Experiments on taste have been carried on by stimu- 
lating the tongue with solutions of 
various degrees of saturation. 

Experiments have sometimes 
been tried with stimuli which are 
undergoing very gradual change; 
no. es that is, instead of requiring the 

observer to compare two clearly 

different intensities of pressure, a given pressure is modified 
gradually. The threshold of discrimination will be differ- 
ent from that which results from a sudden change. The 
same type of experimentation has been worked out for 
gradually varying sounds. The devices for producing 
gradual variations in weight are in principle a form of 
balance upon which weights of greater intensity are gradu- 
ally imposed. Flowing water has l>een used to increase 
the weight of the pressure gradually; the sliding of a weight 
along a counterbalance arm has also been employed. 
For purposes of change in the quality of tones the tone- 
variator of Stern, described on page 111, may be employed. 
The determination of pressure thresholds has been 
touched on in discussing pressure points (page 119). 
An elaborate piece of apparatus for producing and meas- 
uring pressures is shown in Fig. 66. The small ivory 


point St is brought into contact with some part of the skin 
by raising the arm H2. This lever may be raised by some 
form of clockwork which will determine its rate. As the 
lever H2 rises it brings into action the coil spring M z 
gradually increasing the pressure at St. By means of the 
scale the degree of pressure can be directly read. The 
initial position of the lever H 2 can be regulated by the 
screw S and the height of the apparatus after it is clamped 
in position can be adjusted by means of the screw Ms. 
A principle similar to Weber's law can be demonstrated 

FIG. 66 
From the catalogue of Zimmermann, Leipzig, Germany 

for the recognition of the length of lines. If a series of 
lines differing from each other in length is prepared in a 
manner analogous to that described above for the weights, 
experiments by the method of right and wrong cases may 
be carried out. Thus, if a line 10 cm. long is compared 
with a line 10.5 cm. long, and a line 2 cm. long is com- 
pared with one 2.1 cm. long, the methods and results will 
be analogous to those described above for weights. 

Another method may be used for lines which can not 


be employed for weights. In this case the observer may 
be required to draw a line which seems to him to be equal 
to a given line, or he may be required to draw a line which 
is just noticeably shorter or just noticeably longer than 
the given line. Again, the observer may be given a certain 
line and may be asked to cut off that portion of the line 
which seems to him to be equal to a standard. This ex- 
periment has been elaborated as a typical experiment in 
quantitative determination of psychical processes in the 
introduction of the Laboratory Manual and need not be 
worked out further at this point. (See Laboratory Man- 
ual, pp. 3-10.) 



A simple tambour is represented in Fig. 67. A bowl 
B opens at the side into a tube S. The bowl is covered 
over the top by means of a thin rubber membrane 
which is held in position in this form of tambour by a 
tightly fitting metallic ring R. Rubber suitable for tam- 
bours can be secured at any place where dentists secure 
their supply of rubber dam. The rubber is sometimes 
tied around the bowl which is constructed with a groove 
below its edge. In order to insure an air-tight contact, 
the edge of the bowl may be prepared for the reception of 

FIG. 67 
From Professor Porter's catalogue of Harvard Physiological Apparatus 

the rubber membrane by giving it a thin coat of wax. 
Ordinary beeswax serves the purpose very well. The 
rubber should be stretched as little as possible so as to 
leave it free to move with the least possible resistance. 
From the hollow stem S a thick walled rubber tube may 
be carried to any point desired and a second tambour 
may be attached to the other end of the conducting tube, 
the two acting in harmony, one to receive, the other to 
record the movement. In order to avoid loss of energy 
through the elasticity of rubber tubing, it is advantageous 



where the connection is a long one, to insert a glass tube 
in place of the rubber wherever possible. Any pressure 
which is exerted on the rubber surface of one of the tam- 
bours will force air into the other tambour and will pro- 
duce a movement in the rubber of this second tambour. 
The advantages of this arrangement are that movements 
may be taken on one tambour at any point desired and 
may be conveyed to a second point where they can be 

FIG. 68 
From the catalogue of C. H. Stocking Co., Chicago 

recorded on suitable surfaces prepared to receive the 

A record of the movement of a tambour surface can be 
made by means of a lever attached to the surface of the 
rubber as shown in the Fig. 67 at W, P, L. This lever may 
be of an extremely simple form. A wire coil W may be 
wound around a pipe S. At P the wire carries a fulcrum 
and a lever L. This latter is connected with the rubber 
of the tambour by a light angle of metal, C. Any move- 
ment of the rubber surface will be transmitted to the 
lever and magnified at the end of the lever. The degree 


in which the movement is thus magnified will depend on 
the distance between P and (7, and C and the end of the 
lever. PC may be varied by sliding W along S. 

A more satisfactory form of lever is shown in the tam- 
bour represented in Fig. 68. It consists of a post H 
which carries a horizontal bar A which can be set back- 
ward and forward by means of the set-screw X. The 
horizontal rod A is supplied at its end with a fork L and a 
pair of point bearings. The fork can be adjusted up and 
down by the set-screw K which acts against a spring. 
The point bearings in L carry a cross rod to which the 
lever is attached, as represented at F. The lever F is 
fastened to the rubber surface of the tambour by means 
of a light metal rod which has fastened at its bottom a thin 
metallic plate. In some cases the rod which connects the 
lever with the rubber is forked at the top. The lever is 
held in position in this fork by means of a small rubber 
band which passes around the prongs of the fork and over 
the lever. Suitable rubber bands for this purpose may 
be made from a small, pure rubber tube. Such a small, 
pure rubber tube may be cut off in thin sections with a pair 
of scissors, and these sections will supply the band desired. 
The plate at the bottom of the rod is fastened to the rubber 
by means of a drop of wax. This wax can be melted by 
bringing a hot metal bar in contact with the upper sur- 
face of the plate. 

It will be found convenient to mount the tambour on 
a standard which is quite separate from the bowl or from 
the recording lever. This can be done as indicated in 
Fig. 68, where the bowl is connected with a standard by 
means of an arm W. This arm can be mounted on a 
spring S, which is fastened at the end opposite the point 
where it carries W in a clamp by means of which it can be 
fastened to a standard. Connected also with the clamp 


there may be a set-screw 0, which presses against the 
spring S. If now the clamp is held firmly in position and 
the set-screw is operated backward and forward, the whole 
tambour with its recording points can be adjusted back and 
forth so as to be brought nearer the surface on which the 
record is to be made, or removed further from this surface. 
The advantage of such a screw adjustment is that the re- 
cording point can be delicately adjusted to any required 
degree of pressure upon the surface upon which it is to 
record. Such delicate adjustment will be found to be in- 
dispensable if the records are to be taken with as little 
friction as possible. 

Fio. 60 
From the catalogue of Verdin, Paris, France 

A great variety of tambours are supplied by makers. 
Fig. 69 represents a form in which the bowl instead of the 
lever is adjusted. This figure shows very well the plate 
which connects the lever and the rubber. 

It is desirable that the levers used with tambours 
should be as light as possible. Most very light rods have 
the disadvantage of vibrating whenever they move. A 
hollow cylindrical rod is least subject to this objection, and 
fortunately nature has provided light hollow cylindrical 
rods in abundance in straws. The best straw for tam- 
bours is fine, clean rye straw. If this can not be easily se- 
cured, a substitute can be found at almost any florist's 


where sheaves of an Italian grain will be found. This 
imported Italian grain is used as a miniature substitute 
for sheaves of wheat. The straws are fine and straight. 
They are somewhat heavier than rye straws but serve 
very well. 

An important part of a tambour is its recording point. 
This should be at once flexible enough to insure so far as 
possible a reduction of the friction against the record sur- 
face, and it should be rigid enough to insure a clear mark. 
As will be stated later, the form of surface mo'st commonly 
used in making records is a surface of smoked paper. 
Points for tracing on such surfaces may be made of parch- 
ment. The parchment should be cut with sharp scissors 
to a point, and the point should be slightly curled, so as to 
press against the smoked paper surface with the elasticity 
of its curve. Such a parchment point may be fastened 
to the end of the straw by a drop of wax. Or the end of 
the straw may be split, the parchment inserted in the slit 
thus prepared, and the whole bound with a fine silk 
thread. Another very satisfactory point can be made 
by using celluloid, which can be treated in the same 
way as the parchment. Pendulum ribbon makes a good 
metal point. Heavy tinfoil may be used. 

Tambours can be used in various combinations. It 
will be convenient to speak of a tambour supplied with a 
lever and a recording point as a recording tambour. Tam- 
bours which have no levers attached may be variously 
attached to the finger or head or other part of the body 
and will be called, because of their function of receiving 
the movement, receiving tambours. 

It is often convenient in practical work to be able to 
introduce air into a tambour without pulling apart the 
connections. A convenient device for doing this is a short 
metal tube which is inserted in the rubber tube leading 


from receiving tambour to recording tambour. Such a 
tube is represented in Fig. 70. At A is a fine opening 
communicating when open with the outer atmosphere 
and the air system inside of the tambours Under ordi- 
nary circumstances A is kept closed by the small 
stopper BB 1 B", which is pressed down against it by the 
spring S. When it is desired to open A the finger presses 
at B\ 

One of the most fertile sources of annoyance in work- 
ing with tambours is to find that they are not air-tight. 
It is frequently necessary for the student who is to use 
tambours to renew the rubber covering. Indeed, it is 
usually better for him to begin any experiment with an 
entirely new rubber on the tambour. It will be found 

advantageous in every laboratory to have the material 
necessary for setting up tambours at hand in such form 
as to be easily accessible to the student. A box can be 
prepared containing the smaller parts of the tambours, 
a supply of rubber, a supply of strong thread, a pair of 
scissors and beeswax. If the tambour is to be used only 
a short time it is often convenient to fasten the rubber 
tambour by means of a wire instead of thread. A fine 
wire and a pair of pinchers should accordingly be added to 
the equipment. If the wire is drawn firmly around the 
tambour it can be fastened by twisting it up with the 
pinchers. The disadvantage with the wire connection is 
that it very soon cuts through the rubber. 

After securing a recorder, a surface must be provided for 


taking the record. In order that the movements of the 
tambour levers may be fully recorded, the surface must 
offer as little friction as possible and it must be moved for- 
ward so that when one part of the record is completed 
an unused portion of the receiving surface may be substi- 
tuted for the part on which the recorder has been tracing, 
and a new phase of the movement be recorded. The 
movement of the surface also aids in reducing the friction. 
We may, therefore, begin our description with the moving 
device which is known as a kymograph. 

A variety of kymographs are in use. The simplest con- 
sists of a large brass cylinder about 15 cm. in diameter. 
Such a brass cylinder or drum, as it is called, should have 
as smooth a surface as can be provided. It should be 
turned in an accurate lathe so as to be a perfect cylinder, 
and it should be kept polished by means of buffing appar- 
atus so that its surface may at all times be as nearly as 
possible free from irregularities. It is desirable that this 
drum should be rotated at a uniform rate. The prob- 
lem of securing regularity of movement is a problem 
which has resulted in a number of clock-work devices and 
electrical devices. If one wishes the highest degree of ac- 
curacy he will find that all these devices are subject to some 
variations; the effort to drive the drum at a uniform rate 
is therefore at times abandoned, in which case it becomes 
necessary to trace upon the receiving surface a standard 
time-line. Any irregularities in the movement of the drum 
will be indicated by means of this standard time-line, and 
the results of the record to be studied can be measured 
in terms of the standard time-line rather than in terms of 
the movement of the drum. 

Of the devices for driving the drum at a uniform rate, 
the clock-work devices are the simplest. A cheap and 
very satisfactory clock-work device was prepared by Pro- 


fessor Porter of the Harvard Medical School, in the ky- 
mograph which is supplied with the Harvard physiologi- 
cal apparatus. It consists of clock-work shown in the 
accompanying Fig. 71. The drum is carried in a vertical 
position and driven by a friction contact between the foot 
of the drum and the shaft which is connected with the 
clock-work. The whole is regulated with reference to its 
speed by a fan governor shown in 
the figure. The fan governor con- 
sists of a flat piece of metal which, 
in its rotation, strikes the air and 
offers, because of the constancy of 
the pressure of the atmosphere, a 
uniform resistance to the clock- 
work. Fans of various sizes are 
provided with the apparatus. 

More elaborate kymographs are 
made, the best clock-work kymo- 
graph being that which bears the 
name of the physiologist Ludwig. 
This kymograph is supplied by 
Zimmermann, in Leipzig, Germany. 
It is represented in Fig. 72, and 
consists of an elaborate clock-work, 
which by means of the coupling 
and uncoupling of certain of its 
gearings gives a wide range of adjustment of speed. The 
speed of the drum may also be regulated by means of a fan 
governor which consists in this case of a pair of wings 
which spread out against springs because of the centrifugal 
force exerted during rotation. When springs of a given 
strength are placed in position in this governor, the fans 
of the governor tend to spread out to a certain extent, de- 
pending upon the balance between centrifugal force and 

Fio. 71 

From Professor Porter's 
catalogue of Harvard 
Physiological Apparatus 


the tension of the springs, and thus maintain a uniform 
rate of movement in the kymograph. The drum may be 
placed either in the vertical or horizontal position as is 
most convenient for the record. The figure shows at the 
left a tripod standard with a disk and rotating arm and 
contacts to be used in securing regular time intervals for 

FIG. 72 
From the catalogue of Zimmermann, Leipzig, Germany 

the purposes of experimentation with time perception. 
This part of the apparatus will be referred to again in 
connection with the supplementary experiments under 
Exercise XXIII. 

For ordinary laboratory purposes it will be found con- 
venient to use a simple drum not connected with clock- 
work and to drive this with an electric motor. Such an 


arrangement can not be relied upon, it is true, to give as 
uniform speed as clock-work, but it does not need to be 
wound during the course of the experiment and it is a 
very much cheaper arrangement than any which is driven 
by clock-work. It is capable of a great variety of modi- 
fications, and two or three drums of this sort will make 
possible a very large number of combinations for re- 
cording purposes. The drum-shaft should be supplied 
with either a belt-wheel or a cog-gear. The cog is better 
for certain purposes since it provides against the possi- 
bility of any slipping, but for most 
purposes the simple belt is alto- 
gether adequate. By means of a 
series of pulleys and countershafts 
such as are represented in Fig. 73, 
any rate of movement in the drum 
can be secured. Commonly two 
or three intermediate pulleys will 
be placed between the drum-shaft 
and the electric motor. The fan 
motor referred to under Exercise 
III, (page 52) serves very well to 
drive this drum. 

A number of motors have been 

devised which are so regulated that they move at uni- 
form speed. The oldest of these was designed by Helm- 
holtz and is known as the Helmholtz rotation apparatus. 
It consists of the ordinary parts of an electric motor with 
an additional part which regulates the amount of current 
that is supplied to the motor. This regulator consists 
of an arm which can move outward by centrifugal force 
wherever the shaft of the motor is set into rotation with 
sufficient speed. As soon as this centrifugal arm moves 
outward from the shaft it breaks an electric contact and 

FIG. 73 

From the catalogue of Zim- 
tnermann, Leipzig, Germany 


introduces a certain amount of resistance into the electric 
circuit which drives the motor. The reduction of the cur- 
rent tends to reduce the speed of the motor. As soon as 
the speed is reduced the centrifugal arm falls back again 
against the shaft and the extra resistance is cut out. 
The motor once more begins to move more rapidly, the 
centrifugal arm again moves away from the shaft and in-, 
troduces the resistance, and so on. The point of regular 
rotation in this apparatus is reached when the centrifugal 
arm makes a continuous light tapping due to its constant 
movement away from the shaft and back again. Recently 
devices have been worked out for regulating a motor by 
means of a tuning-fork. 

The drum and its motor being provided, the next step 
consists in the preparation of a suitable surface for taking 
the record. A strip of heavy glazed paper, no wider than 
the drum, should be pasted around the drum. In doing 
this care should be taken to bring the paper as smoothly 
in contact with the drum at all points as possible, and to 
paste it under such tension that it shall be held firmly in 
position against the metallic surface. The paste should 
not be applied to the surface of the drum but along the line 
of contact of the two ends of the paper. Instead of pasting 
a sheet of paper on a single drum, it is very convenient, 
especially if a long record is to be taken, to carry a belt 
of paper between two drums, as indicated in Fig. 74. If 
a drum A is driven by means of clock-work or by means 
of an electric motor, it will drive the drum B which does 
not need to be connected with the driving device except 
through the belt of paper C. In practical operation this 
belt of paper can be made as long as the strength of the 
paper will permit. Belts of paper 70 feet long have been 
utilized, although care must be taken in the use of such a 
long belt not to bring too great a strain upon it at any 


part of the belt during the process of taking or fixing the 
record. The adjustment of the belt of paper on these 

FIG. 74 

two drums requires some manipulation. In the first place, 
the two drums must be in alinement and the ends of the 


paper must be carefully pasted so as to make a true joint 
in the paper. The two drums will probably need to be 
adjusted after being set up. They may be brought into 
approximate position and the belt slowly rotated. If 
the belt tends to run off the drums, the drums should be 
readjusted with reference to each other on the following 
principles: If the belt tends to run to the right or left on 
the remote drum, the near drum should be set in a direc- 
tion opposite to that in which the paper tends to run. 
Thus, if the paper runs toward the right of the remote 
drum, the drum near at hand should be moved gradually 
toward the left. If the movement is made gradually, the 
paper can usually be brought into a position of equilibrium 
very readily. Rapid movements are unfavorable to the 
adjustment. If instead of running off of the remote drum, 
the paper tends to run to the right or left on the drum near 
at hand, this tendency may be corrected by moving one or 
the other end of the drum toward or away from the remote 
drum. The belt travels in the direction of least tension, so 
that if the belt tends to run off at the right side of the drum 
near at hand, it indicates that there is less tension on the 
right-hand side of the belt than on the left-hand side. 
The right-hand end of the drum should, therefore, be drawn 
away from the more remote drum so as to increase the ten- 
sion of the belt on the right side. With a little practice 
the experimenter can learn to manipulate the belt easily, 
and to adjust it so that it will run smoothly and in a given 
part of the two drums. The most important condition 
for this desirable result is the careful pasting of the strip 
of paper at the outset of the experiment. 

When the belt is adjusted or the paper pasted upon the 
kymograph, the next step is to lay on a coat of lampblack. 
A variety of forms of combustion can be employed for pro- 
ducing a layer of lampblack. In every case the heat 


should be applied to the paper only where the paper is in 
contact with the drum. The metal drum serves to con- 
duct away the heat and prevents the paper from burning. 

The finest grade of lampblack can be produced by burn- 
ing camphor under the paper and rotating the drum at a 
considerable speed, carrying the paper through the smoke. 
Camphor is more expensive than some of the other sub- 
stances which can be employed. Ordinary illuminating 
gas gives a very good fine layer of lampblack. The form 
of flame which is most advantageous is one which gives the 
least possibility of complete combustion of the illuminat- 
ing gas. A wide flame supplied with gas through large 
holes is the most desirable flame for this purpose. The 
drum should be rotated so as to draw the flame underneath 
the drum, and the flame should be held in such a position 
that it is in contact with the paper only at the red, upper 
part of the flame. Coarser and less desirable forms of 
lampblack may be produced by burning under the paper 
kerosene or even turpentine. The latter produces very 
rapidly a thick layer of lampblack, but it is coarse and 
therefore not so advantageous for delicate work as some 
of the other forms mentioned. Automatic devices for 
smoking strips of paper have from time to time been de- 
scribed; they are all very complicated and can be dispensed 
with, especially if one uses belts instead of single sheets of 

When the tambour point has been allowed to trace on 
this lampblack surface and the record has thus been se- 
cured, it may be made permanent by shellacing or varnish- 
ing the lampblack surface. The most convenient device for 
shellacing or varnishing consists in the apparatus repre- 
sented in Fig. 75. A tray T is mounted in a table S and 
may be covered when not in use by a cover. At the bot- 
tom of the tray a rubber tube is carried, as indicated by 


R, to a bottle which contains the liquid varnish or shellac. 
This bottle should be suspended by means of a string 
which passes through the pulley P and is fastened at the 
bottom to a pedal M. When it is desired to shellac a strip 
of paper, the foot is placed upon the pedal M, and the 
bottle containing the varnish or shellac is raised to a level 
higher than the tray T, and the varnish runs into the tray 
T. As soon as the foot is removed from the pedal, the 

FIG. 75 

bottle sinks to a level lower than the tray and the varnish 
runs back into the bottle. It is not necessary to immerse 
the paper taken from the kymograph in the varnish or 
shellac. It is enough that the varnish or shellac be brought 
into contact with the back or unsmoked surface of the 
record. This is accomplished by drawing the strip of 
paper, after it is removed from the drum, through the tray 
so that its unsmoked surface comes in contact with the 


shellac. The shellaced paper may be hung in a holder 
above the tray and allowed to drip into the tray, thus 
saving all the unused fluid. If it is desired to keep the 
records as permanent records, it will be necessary to pro- 
vide a good quality of shellac. If the record is merely for 
temporary use any thin transparent varnish will serve the 

If the record is made on a long belt as shown in Fig. 74, 
the simplest method of fixing it is to shellac the belt on the 
unsmoked surface while it is still on the drums. A bottle 

of shellac is provided with a 
bent tube in the cork, as 
shown in Fig. 76. From this 
a small stream is poured on 
the back or unsmoked sur- 
face of the long strip and 
spread by means of a large 
brush so as to cover all parts 
of the paper. If this process 
76 of shellacing is undertaken 

with some deliberation there 

will be no danger of the paper being torn or of covering 
the drums with wet shellac. If the work must be done 
hurriedly, the drums are likely to be covered with the 
shellac and it will be necessary to clean the drums from 
time to time by buffing them. It will be especially neces- 
sary to guard against the tearing of the paper, which is, 
of course, very much weaker when wet with the shellac 
than it is when dry. 

The record which is secured by allowing a tambour 
point to trace upon a moving surface will depend for its 
length upon the length of the movement recorded and the 
rate of movement of the record surface. If this surface 
is moving rapidly, the record of a given movement will be 


long drawn out; if the surface moves slowly, the record will 
be much condensed. If the surface moves at a uniform rate, 
the various parts of the record will be directly comparable; 
if the surface moves at an irregular rate, comparison of the 
various parts of the record can not be directly made. It 
is often desirable to state the results from records obtained 
in this way in absolute terms and to provide, as indicated 
in an earlier paragraph, against the possible irregularities 
in the movement of the receiving surface. In such cases, 
a time-line from a vibrator of known rate may be traced 
on the paper along with the tambour record. 
The simplest method of securing a standard time-line 


FIG. 77 
From the catalogue of C. H. Stoelting Co., Chicago 

is to use a Jaquet chronometer manufactured by Verdin, 
in Paris. This is a small stop-clock with a lever attach- 
ment which moves every second or every fifth of a second 
according as it is set for the one interval or the other. 
The lever can be made to trace directly on a kymograph. 
Another method of securing a time record is to attach 
a point of pendulum ribbon, similar to that described in 
connection with the tambour, to a tuning-fork or other 
vibrator and hold this point during the vibration of the 
fork or rod against the record surface. If a very short 
unit of time is desired, a 100-vibration fork or a 500- 
vibration fork may be used. The difficulty with such 
a contrivance is that the time record is very short, or else 


the experimenter must devote himself to keeping the fork 
in motion. A device for keeping the fork in motion is pro- 
vided in the electric fork, which operates on the same 
principle as the vibrator described on page 99. Such a 
fork is represented in Fig. 77. It consists of an ordinary 
fork mounted on a frame and having between its prongs an 
electromagnet. The current for this electromagnet enters 
through the fork and passes from the fork to the magnet, 
whenever contact is made between the spring S and the 
plate, which can be adjusted by means of the screw P. 
S and P will be in contact whenever the prongs of the fork 
vibrate outward. As soon as S and P come together, the 
current passes from the fork through the magnet and back 
to the battery. The magnet will now attract the prongs 
of the fork inward with a strong impulse. The rate at 
which the prongs of the fork respond to the attraction of 
the magnet will be determined by the fork's natural rate of 
vibration. The energy for the movement will be supplied 
by the magnet. As soon as the prongs of the fork vibrate 
inward in response to the attraction of the magnet, the 
contact between S and P will be broken, the electromagnet 
will cease to attract the fork, and by its own elasticity the 
fork will tend to vibrate outward again. In vibrating 
outward the contact at S and P is established again and the 
process is indefinitely repeated. 

An electric fork has two uses; it may be used directly to 
make a tracing on a kymograph or it may be used indi- 
rectly to make and break an electric current, this current 
in turn being used to make the record on the drum. 

If the fork is to be used directly, a point is attached to 
the end of one of the prongs, and the fork is held in contact 
with the drum. A convenient means of holding the fork 
is the clamp shown in Fig. 78. The shaft R passes through 
a hole drilled through the base on which the electric fork 


is mounted. The fork is held firmly on this shaft by 
means of the screw S. This rod R is clamped by the screw 
/ to a second rod P, which is hollow and fits over the third 

rod T, which is held by the 
standard. From P and T 
two arms project, as shown 
in and Q. Between these 
arms is the set-screw N which 
controls the distance between 
them. T is clamped to a 
holder, the fork is fastened at 
FIG. 78 RS , and by means of N the 

From the " Studies from the Yale Psy- + f j.^ frvrL- rna\r KA 

choiogicai Laboratory," Vol. iv position ot me torK may oe 

nicely adjusted up or down 

so as to regulate the pressure of its point against the 
smoked paper. 

If the fork is to be used indirectly, a marker is required. 
A simple marker is represented in Fig. 79. An electro- 
magnet ra is connected in series with the electric tuning- 
fork. A current passes through ra every time a contact 
is made at the fork. Directly in front of the metallic 
core e of the magnet ra, is placed a spring point a. The 
marker a is held by its own elasticity away from the magnet, 
but not far enough away to be beyond the range of attrac- 
tion when the current passes 
through the magnet. As a re- 
sult, the marker a moves toward 
the magnet whenever a current 
passes through the coil and os- 
cillates in the opposite direction FlQ 79 *" 
when the current in the coil is 

interrupted . Very often the spring which draws the marking 
point away from the magnet is a coil spring. The moving 
point a is made to trace on the smoked paper surface. 


Various forms of markers have been devised. In gen- 
eral if one's equipment is limited, it is better to procure a 
small marker. This can be used for either a slow or a 
rapid record, whereas a heavy marker has too great in- 
ertia for rapid vibrations. 

The principle of the electric tuning-fork and of the in- 
terrupter described on page 99 may be elaborated so as 
to give any desired rate of motion and any desired strength 

Fio. 80 
From the catalogue of Zimmermann, Leipzig, Germany 

of current. One of the most elaborate instruments con- 
structed on this principle is the Kronecker interrupter. 
This interrupter is supplied with mercury contacts for 
making and breaking the current. It is also supplied by 
means of an additional pair of contacts with the means of 
making and breaking a current entirely separate from 
that which drives the vibrating rod. It is supplied with 
a system of tubes by means of which the mercury con- 
tacts are kept clean by a flowing stream of water that 


passes over these contacts. This stream of water is sup- 
plied from a small reservoir which stands a little above 
the mercury contacts. From this reservoir the water is 
conducted by tubes directly across the mercury. The pla- 
tinum needles pass through the water and into the mercury, 
the contact not being completed until the needles reach 
the mercury. If it is desired to secure intervals of time 
of a second or longer, a pendulum may conveniently be 
used. A contact pendulum manufactured for this purpose 
and giving series of contacts for any interval from one 
half of a second to a minute is shown in Fig. 80. With 
each oscillation of the pendulum the dial is advanced one 
degree. An outer series of metallic posts is so arranged 
that with each movement of the dial a post comes in con- 
tact with the catch. Other series of posts on the dial are 
separated by longer intervals. The catch may be adjusted 
so as to connect with any series desired. The catch con- 
trols an electric current which is thus regulated by the 
clock for any long interval desired. A simple contact may 
be arranged with any pendulum by allowing it to sweep 
through a meniscus of mercury. In such a case one wire 
should be connected with the pendulum, the other with 
the mercury. 

It is sometimes possible and desirable with markers 
which record long intervals and consequently move slowly, 
to make a record by means of a pencil or pen on a paper 
which is not smoked. The pencil used for such purposes 
is mounted at the end of a flexible metal strip which presses 
it against the paper and is at the same time sufficiently 
elastic to avoid excessive friction. A pen may be similarly 
mounted. The best form of pen is one which feeds out 
the ink through a capillary opening. A glass tube drawn 
to a fine point makes a very good pen. Metallic pens 
with holes of very small caliber can also be used. 


The records which are made by a tambour can be used 
not merely for the purpose of determining the duration of 
movements but also for the purpose of studying the in- 
tensity and form of these movements. The intensity and 
form are, however, by no means as directly recorded as 
the rate, for the tension of the tambour rubbers increases 
rapidly after the rubber has been stretched even very 
slightly. The last part of the movement of a tambour 
lever represents not merely the energy necessary to lift 
the lever itself, but in addition the energy necessary to 
stretch the rubber. For this reason the quantitative use 
of a tambour record to show the amount of movement in- 


volves complex corrections. In a general way a higher 
movement means a greater amount of energy of move- 
ment, but the amount of energy expended is not directly 
proportional to the movement of the tambour recorder. 
A record is presented in Fig. 81 showing a regular time- 
line from a 100-vibration fork. A second line shows the 
tambour record of an upward and downward finger move- 
ment which is executed as rapidly as possible. It will 
be seen that the successive finger movements differ from 
each other in intensity and in regularity of form as well 
as duration. 

It remains to mention in connection with the kymo- 
graph and its accessories, standards for carrying the vari- 


ous recording pieces. The simplest standards consist 
merely of heavy metallic bases into which are set rods. 
The best bases for such standards are heavy tripods. 
More convenient than these simple standards is one which 
can be raised or lowered by means of a screw. Such screw 
standards may be connected with kymograph drums so 
as to move the tracing points automatically as the drum 



The simplest contrivance for securing records of the 
rate of the heart-beat consists in a recording tambour, such 
as was described under the last exercise, and a rubber tube 
connecting this recording tambour with a thistle tube such 
as is shown in Fig. 82. The large end of the thistle tube 
should be placed against the skin of the neck just over 
the carotid artery and should be held in position by rods 
and clamps fastened to a table or portable base. The 
skin of the neck serves as a cover for the thistle tube, which 
is thus converted into a tambour, and the pulse of the 
carotid artery will at intervals compress the air in the 

thistle tube, with the result 
that the recording tambour 
will be set in motion at the 
, same rate as the pulse. The 

difficulty with this arrange- 
ment is that any movement of 

the reactor's head and any motion of swallowing or un- 
usual breathing will change the pressure of the air within 
the thistle tube and complicate the pulse record. The 
first difficulty can be very largely overcome by fixing the 
head in the head-rest already described under Exercise 
II, page 31. No easy means of overcoming the diffi- 
culty that arises from swallowing can be provided. It is 
better for this reason to utilize the pulse in the radial 

The radial pulse is not strong enough and the skin of the 



wrist is not smooth enough to allow the use of a thistle tube, 
and a special contrivance must be provided for recording 
this pulse. The simplest contrivance that can be used is a 
cork fastened to an ordinary recording tambour. If one 
end of the cork is fastened to the rubber surface of the 
tambour by a drop of wax, and the whole tambour is held 
in a clamp in such a position that the free end of the cork 

FIG. 83 

presses against the radial artery, a record may be taken 
with a recording tambour which will correspond in rate 
and in intensity to the rate and intensity of the movement 
induced in the receiving tambour by the pulse. 

Fig. 83 shows a more elaborate contrivance for record- 
ing the radial pulse. A spring $, with a small plate P, is 
brought by means of a frame F into contact with the radial 
artery at the point P. Bands are provided which will 
hold the frame firmly against the arm. These bands are 
not represented in the figure. Passing through a wire 


hook L connected with the spring is a small rod which 
is fastened perpendicularly to the long bar W. At the 
other end of H 7 is a long arm K , which serves to magnify the 
movements communicated by the spring at L. This en- 
larging lever may be made to carry a point which will trace 
directly upon a smoked surface, or it may be connected 
with other enlarging devices before the tracing is made. 
A very effective enlarging device is represented in Fig. 
83. The end of the lever K is made into a loop as shown 
at N. Through this is allowed to pass a rod which is 
weighted on a short perpendicular arm by a small weight X. 
This weighted rod has its fulcrum at R. 
At the end of its lower long arm J is a 
loose joint with a tracing point 0. The 
weight draws the rod RJ into the posi- 
tion of rest against the loop N. If the 
loop N is moved, the rod will be moved 
with it and the record can be taken at 
0. The result will be that the move- 
ments of the lever K are a second time 
enlarged by this additional leverage. 
In the use of such an apparatus, which is technically 
known as a sphygmograph, care must be taken to have 
the relation between the arm and the recording surface 
perfectly constant. If the arm is allowed to move in any 
degree, the tracing at the end of the pointer will be affected 
by the arm movement as well as by the movement of the 
pulse, and a source of error will thus be introduced in the 

A simple device for holding the arm in a fixed position 
consists of an arm-rest made up of two U-shaped brackets 
shown in Fig. 83a, through which the arm is allowed to 
pass and lie in a comfortable position. The hand should 
clasp firmly the holder, which can be adjusted by means of 


a rod to any desired level. A sphygmograph attached to 
the arm held in such a rest as this can be adjusted so as 

FIG. 84 
From the catalogue of Zimmermann, Leipzig, Germany 

to trace upon a drum, which has been set in the proper 
position with reference to the arm-rest. 

There are other sphygmographs. In some of them the 
movement of the spring which presses against the pulse is 
communicated to a tambour and thus recorded on a 
kymograph. Such a sphygmograph is shown in Fig. 84. 
P presses against the artery with a pressure controlled by 

FIG. 85 

the spring M . As P moves up and down it acts upon the 
tambour shown in the figure. The frame is to hold the 
apparatus on the arm. 



Fig. 85 shows a typical record from the sphygmograph. 
The following table shows the change in heart-beats 
resulting from the various forms of excitation : 

1. Multiply 31 by 17. 

2. Multiply 27 by 13. 

3. Strong disagreeable odor. 

4. Pleasant odor. 

Average number of 
heart-beats in 5 se- 
lected periods of 10 
seconds during 2 min- 
utes preceding stim- 

Average number of 
heart-beats in each 
10 seconds during 30 
seconds immediately 
after stimulus 

Average number of 
heart-beats in 5 se- 
lected periods of 10 
seconds during 2 min- 
utes following stim- 

M. V. 












A second type of apparatus for securing records of vaso- 
motor changes is the plethysmograph. This is made to 
enclose some part of the body and record changes in the 
volume of a part of the body which it encloses. 

The simplest plethysmograph is made for the finger. 
It consists in a small glass cylinder. This cylinder has 
a thin rubber hood extending into it to receive the finger. 
This rubber hood is fastened to the edge of the cylinder 
so that the whole makes what may be called a hollow 
tambour. The finger of the reactor can now be inserted 
in the rubber hood, and the finger and cylinder are held 
firmly in position by means of the necessary arm-rests 


and clamps. If the volume of the finger increases or de- 
creases, as it will with each pulse-beat, or if it undergoes 
more gradual changes through the withdrawal of the blood, 
or through an excessive flow of blood to the finger, the air 
between the wall of the cylinder and the rubber hood will 
be driven out or drawn in each time there is an increase 
or decrease in the volume of the finger. If a recording 
tambour is connected with the cylinder, changes in the 
volume of the finger will be recorded. 

The principle employed in this finger plethysmograph 

FIG. 86 
From the catalogue of Zimmennann, Leipzig, Germany 

can be applied to a larger apparatus for the arm. Fig. 
86 represents an arm plethysmograph supplied by Zim- 
mermann. The arm is inserted in the large cylinder and 
held in position by the elbow rest. The cylinder contains 
a large rubber sleeve, and the space between this sleeve 
and the walls of the cylinder is filled with water. The 
water rises into the glass tube extending vertically out of 
the cylinder, and from this glass tube connections are car- 
ried to a recording tambour. Rubber sleeves for this 
plethysmograph can be made at any rubber establishment 
by using a mold of proper size made of polished wood or, 


still better, of glass. The objection to this form of plethys- 
mograph is that the rubber sleeve or sack does not fit 
closely around the hand, especially at the end, and error 
is likely to be introduced into the record through involun- 
tary movements of the fingers. Such movements result 
in changes in the volume of the rubber sack, which are in 
no way due to changes in the volume of the arm. One 
method of reducing this error to its minimum is to close 
the hand firmly about some solid object and keep it clenched 
inside of the rubber sleeve. 

A still better device consists in eliminating altogether 
the closed end of the rubber sleeve. An outer metallic 
cylinder, open at both ends, is drawn over the arm. This 
cylinder is supplied with a rubber sack which is made 
fast at the ends of this tube. The hand thus extends 
freely beyond the rubber sack. If, now, water is allowed 
to fill the cylinder between the rubber sack and the wall 
of the cylinder and to rise for a short distance in a vertical 
tube, similar to that shown in Fig. 86, a record can be 
taken as in the other cases by means of a recording tam- 
bour. This form of plethysmograph has the advantage 
over the finger plethysmograph of including a larger part 
of the body and, therefore, insuring a greater change in 
volume. It has a decided advantage over the Zimmer- 
mann plethysmograph in that no involuntary hand move- 
ments will change the volume of the arm. 

A form of apparatus which has recently been described 
by Professor Henderson could be used to great advantage 
in experiments on the relation of circulation to conscious 
processes. The apparatus, called a recoil table, consists 
of a table on which the reactor lies. This table is sus- 
pended or supported on flexible supports so that it may 
readily swing through a small distance. Levers are con- 
nected with the table so as to record any movement. 


Whenever the heart of a reactor drives the blood into the 
aorta there will be a recoil of the body, in accordance with 
the familiar physical principle of action and reaction, in 
a direction opposite to that in which the heart forced the 
blood. This recoil of the body is communicated to the 
table on which the reactor lies and through the leverage is 
recorded on the kymograph. 

A form of apparatus required in the third accessory ex- 
periment of this exercise for measuring changes in the 
rate of intensity or respiration is known as a pneumograph. 
A very satisfactory pneumograph is represented in Fig. 

FIG. 87 

87. It consists of a coiled spring C twenty-five to forty 
centimeters in length, covered with a light rubber tube 
RR, which is entirely closed at one end and at the other end 
is fastened firmly around a small metallic tube which per- 
mits connection with a rubber tube leading to a recording 
tambour. From the ends of the coiled spring A and B 
extends a chain which is fastened about the thorax of the 
reactor. When the pneumograph is in position, the coiled 
spring C should be under slight tension. Any change in 
the volume of the thorax will result in a change in the ex- 
pansion of the coiled spring C, and as a result the volume 


of the air column within the rubber tube R will be modi- 
fied and the changes in this air column will be recorded at 
the recording tambour. By using more than one pneu- 
mograph, tracings can be made for different parts of the 
thorax during a single act of respiration. By using at the 
same time a pneumograph and a plethysmograph or 
sphygmograph, parallel tracings can be made for respira- 
tion and the heart action. 

For reasons which were pointed out in the earlier dis- 
cussion of tambour records (page 158), the records from 
sphygmographs, pneumographs, and plethysmographs can 
not be regarded as direct quantitative records of the amount 
of movement of the pulse or lungs. The recoil table has 
a distinct advantage in this respect, in that the amount of 
movement of the table is directly proportional to the 
amount of blood moved at each heart-beat. The records 
from sphygmographs, pneumographs, and plethysmographs 
may, however, be treated quantitatively so far as the rate 
and general intensity of the movement is concerned. 

For a study of changes in the size of the pupil of the eye 
the best method is to photograph the eye in a good light. 
The light for this purpose may be sunlight reduced in 
intensity by passing it through a screen of blue glass, the 
screen being of any thickness necessary to prevent dis- 
comfort on the part of the reactor. 

If the eyes are not photographed, measurements may 
be taken by observing the eyes of the reactor through a 
reading telescope placed at some distance in front of the 



The steadiness of the finger can be measured by al- 
lowing it to rest upon the rubber surface of a receiving 
tambour, any movement of this tambour being carried to 
the recording tambour and there traced upon smoked 
paper. The only precaution required is that the pressure 
shall not be so great as to bring the finger into contact with 
the bottom of the tambour, nor so light as to make no im- 
pression on the rubber. Another somewhat simpler de- 
vice is to allow the finger to rest on one end of a lever 
which is held in position by means of a fulcrum and traces 
at the end opposite that which is in contact with the 
finger, on a kymograph. Slight movements of the finger 
can be magnified in this way as much as desired. In like 
fashion, head movements may be recorded by connecting 
the head by means of a rigid rod with a receiving tam- 
bour or with a tracer which marks directly on a recording 

Various forms of planchettes have been devised. The 
simplest consists of a board hung from the ceiling of the 
room by a long string or wire. A board thus suspended 
by a long string can be moved very freely, and its move- 
ment will be practically in a straight line through any ordi- 
nary distance such as would be required for the experi- 
ment. Another device for furnishing a recorder for hand 
or arm, which can be moved with very little friction, con- 
sists of a glass plate supported on a second glass by large 
balls, such as are used for ball bearings. After securing by 



one of these means a surface which can be moved with 
relatively little friction, it is necessary to provide a tracing 
point. This tracing point should add as little as possible 
to the friction of the movement. In some cases a pointer 
has been allowed to trace upon smoked paper. The paper 
is smoked on a kymograph and is spread out on a flat 
surface under the pointer. The pointer may be of the 
simple sort described in. discussing tracing points for tam- 
bours or it may be a stylus especially designed for use 
with the planchette. A very convenient form of stylus 
consists of a rod which passes freely through a glass or 
metal tube. Such a stylus always rests on the receiving 
surface with its own weight. The metallic or glass 
tube through which it passes is fastened to the 
planchette. As the planchette moves backward 
and forward the pointer will trace its movements 
on the smoked paper. The smoked surface may 
be dispensed with if the stylus is made in the 
form of an ink pen. For this latter purpose a 
special pen-point must be provided which will 
trace equally well in any direction. The ordinary pen-point 
will not serve because it traces freely only in one direction. 
Fig. 88 represents a form of pen-point which answers the 
purpose. It consists of a small metal rod turned down 
to a point into which a groove has been sawed as repre- 
sented at G. The two points P and P r have been bent 
together so that they are nearly in contact. A hole which 
serves as a reservoir for ink is drilled at the upper end of 
the slit between the points. The size of this reservoir is 
determined by trial. If it is too large the ink is forced out 
too rapidly. If it is too small it requires very frequent re- 
filling. Such a pen as this should be prepared for making 
the record by being filled with some thin fluid ink. The 
best ink for the purpose consists of a water solution of one 


of the aniline dyes. Such a pen as this will mark freely 
in any direction, and the amount of ink it will carry is quite 
adequate for an ordinary record. 


A number of very striking records of involuntary hand 
movements are reported in Jastrow's Fact and Fable in 
Psychology, pp. 307-336. One 'of these figures is repro- 
duced in the author's Psychology, General Introduction, 
page 187. 


The apparatus necessary for the supplementary experi- 
ments suggested in connection with this exercise is usually 
some form of tambour. Thus, if a tambour is pressed 
against the cartilages of the 
larynx, a record will be se- 
cured of the movements of 
the muscles of the larynx. 
Special frames have been de- 
vised for holding the tambour 
in position upon the larynx 
or against the other parts 
of the neck. One such, de- 
vised by Rousselot for re- 
cording upward and down- 
ward movements of the larynx, is represented in Fig. 89. 

In experiments with the tongue, it is desirable to pro- 
vide some means of inserting the whole receiving tambour 
in the mouth. For this purpose rubber bulbs have been 
devised of the form shown in Fig. 90. These can be taken 
into the mouth and can be held between the tongue and 
he roof of the mouth or between the tongue and the teeth. 

Fio. 00 

From Rouaaelot's " Principes 
Phon^tique Experimentale " 


Such rubber bulbs are in essence tambours in which all of 
the surfaces are more or less flexible. Any compression of 
the bulb will drive the air out into the recording tambour. 

Dynamometers for measuring 
the strength of movements are 
of various forms. One of the 
simplest forms consists in a 
double spring and recording 
dial shown in Fig. 91. This is 
grasped in the hand and pressed 
together as the hand closes. 

Instead of taking merely the 
record of a single compression, 
devices may be attached to this dynamometer for ad- 
ding up the movements made in a number of successive 
efforts; or instead of a recording device directly con- 
nected with the dynamometer, a pointer may be carried 
to a recording surface and the amount of work done may 
be recorded by tracings on smoked paper. 

It will be found by the use of such apparatus that the 
greater the stimulus affecting the organs of sense, the 
greater will be the amount of work which can be done. 

An illusion which serves very well to illustrate the re- 
lation between perception and change in muscular tension 
appears whenever one lifts two blocks which are of the 
same objective weight but are of 
unequal size. Quantitative 
methods of determining the 
amount of this illusion can be 
worked out by presenting to the 
observer a series of weights in- 
termediate in size between the large and small weight. 
The intermediate series should range in objective weight 
by short stages from 30% less than the weights to be 

Fio. 91 


compared to 30% above. Ten to fifteen weights of this 
sort being presented to the observer, he should be al- 
lowed to select those of the intermediate weights which 
seem to him to be equal to each of the two weights which 
constitute the illusion. In one case, for example, where 
the primary weights were 55 grams, it was found that 
the large weight seemed to the observer to be equal to 
45 grams, whereas the small weight seemed to be equal to 
65 grams. The illusion must therefore be equal in this 

FIG. 92 

case to 20 grams, which is somewhat more than 36% of 
the objective weight. 

A more productive method of dealing with this illu- 
sion is to attach to each of the two weights, levers which 
will record the way in which the observer lifts the weights 
when he judges of their intensity. Fig. 92 shows an elab- 
orate apparatus for recording the movements of the ob- 
server in testing these weights. A large weight A stands 
upon a platform P. By the side of A is a shelf S upon 


which is placed a small weight B. The purpose of this 
shelf is to bring the small weight B to the same level as 
the large weight A. Each weight is supplied with a screw- 
eye HH t so that when the subject lifts the weights he may 
have the same type of contact in both hands. Strings 
pass down, as shown on one side, from each of these 
boxes over pulleys YY. These strings are held in posi- 
tion by weights not shown in the figure but fastened at 
the ends of the thread at WW. The threads are con- 
nected with compound levers LL which, as seen in the 
figure, trace lines upon the paper, giving a record as indi- 
cated at R. This record makes a comparison between the 
two movements very easy. When both boxes are at rest 
the two levers make straight, parallel lines as indicated 
in the drawing. When they are lifted they describe 
curves which in form and in time indicate in detail the 
mode of lifting the weights. Typical records from this 
apparatus are described and discussed by Mr. Loom is in 
the Yale Psychological Studies, New Series, Vol. I, No. 2, 
pages 334-348. 


FIG. 93 


The simplest device for securing a record of rapid 
movements of the fingers is a contact key such as that 
represented in Fig. 93. This spring key has a metallic 
connection at the wires W l and W 2 
with a battery which supplies a cur- 
rent and with a marker which records 
on a kymograph whenever a contact 
is made at the two points on P and 
R. The distance between P and R 
should be made as small as possible 
so that the electrical contact may be 
made as soon as the fingers press 
down upon the plate R. The record of the movement 
of the fingers obtained at the marker should be com- 
pared with the standard time-line derived from some 
vibrator, either a tuning-fork or a vibrating electric rod. 

The apparatus for measuring the rate at which plain 
lines are drawn by a reactor, consists in a wide strip of 
paper which is drawn forward by a kymograph under a 
platform upon which the hand of the reactor is supported. 
The rate of the paper should be indicated by means of a 
time-line which is traced on the edge of the paper by means 
of a pencil or ink marker such as that described on page 
155. If, while the paper is moving, the reactor draws on 
the paper with a pencil lines perpendicular to the direction 
in which the paper is traveling, each line which is drawn 
on the paper will be drawn out into a record which is in 



its form a resultant of two movements the movement 
of the paper, which is of known rate, and the move- 
ment of the hand, which is to be measured. A determi- 
nation of the rate of the hand movement from this record 
is very simple so long as the hand movement is strictly 
perpendicular to the line in which the paper is traveling. 
If the movement of the hand is in some other direction, 
it will be necessary to complicate the arrangements thus 
far described by inserting above the moving paper, first, 
a piece of carbon paper or a typewriter ribbon, and second, 
a piece of paper on which the line is to be drawn. The 
paper on which the line is to be drawn, should in this case 
be held firmly in a fixed position. The moving paper 
should pass beneath the carbon paper and should be 
marked by a time marker to indicate the rate of its move- 
ment. If now a line is drawn freely upon the upper fixed 
paper, a tracing will be produced by means of the carbon 
paper or typewriter ribbon on the moving paper. The 
line thus traced on the moving paper will, as before, be 
the resultant of the hand movement and the rate of the 
movement of the paper. The line drawn on the fixed 
paper will show the length and direction of the hand 
movement. Comparison between the line on the fixed 
paper and the line on the moving paper can now be easily 
instituted, and the rate of the hand movement can be de- 

An elaborate apparatus constructed on the principle 
just described is reported as follows by Mr. Freeman in 
the Yale Psychological Studies, Vol. I, No. 2, pages 

"Fig. 94 gives a general view of the apparatus. Two bars ex- 
tending from the metal base A support the roll of paper and the spool 
of ribbon from which the strips B and F are unrolled. B and F pass 
across plate A to the drum C and the spool S, the latter being sup- 


ported by the post Q and one not shown in the figure. The drum 
and spool are driven through spur-gear connections by the shaft G, 
which is in turn connected with a driving shaft. The apparatus 
is coupled into the driving shaft and uncoupled by a friction clutch 

FIG. 94 

of the type shown in Fig. 95. This consists of a large balance wheel 
F which is driven by an electric motor. The heavy balance wheel 
is necessary in order to maintain uniform motion when the appa- 
ratus is thrown into gearing. The balance wheel carries a hollow 
cone into which a solid cone (W} may be firmly set from above. The 
solid cone is in turn connected with the shafting (S) which drives 
the drum. By means of the handle (H), which holds the solid cone 
in a ball-bearing collar, the 
solid cone may be lifted out 
of the hollow cone, when the 
shaft (S) will be uncoupled 
from the driving wheel. On 
the other hand, when the 
solid cone is set firmly into 
the hollow cone, the shaft (S) 
will immediately be set in 
operation at the full speed of 
the driving wheel. The up- 
ward and downward movement of the shaft is taken up by a slot 
device at the upper end of the shaft. By means of this clutch the 
apparatus shown in Fig. 94 can be set in motion at full speed, and 
it can also be instantly set free when the record is complete. 

FIG. 95 


" In order to hold in position the primary sheet of paper on which 
the reactor writes and to support the hand, a plate, H, is placed 
over the primary sheet. A rectangular opening, N t is made in this 
plate to expose a writing surface on the primary sheet. The plate, 
H, is hinged at the back of the main base by two bars, so that it may 
be raised to insert the paper. Fig. 96 shows it raised from the base. 


FIG. 06 

Two small pins, 0, O, pierce the primary sheet of paper and fit in 
the hole R t and one not shown in the figure, and keep the primary 
sheet from slipping when the strip and ribbon pass beneath it. These 
pins are above and below the moving strip. In order to get an even 
writing surface the plate is set into the main base, so as to lie flush 
with the general surface, and is held down by a screw, R, Fig. 94. 
The moving strip of paper and ribbon are also set below the surface 


in a channel which is cut in the main base. Two guides, T, T, on 
each side of the moving strip keep it straight. The upper ones are 
slightly adjustable, so as to suit minor differences in the width of the 

" In order to obtain a record of the relative position of the primary 
sheet and the moving strip, two pencil points are set through holes 
in the hinged plate, H. These pencil points make two dots upon 
the primary sheet and two lines on the strip. The points are shown 
in Fig. 94, X, X, and the holes through which they project in Fig. 96, 
E, E. The points are set on two flat springs and are adjustable with 
screws, so that they may be set against the paper with varying de- 
grees of pressure. 

"Since the speed of movement of the strip is not perfectly uniform, 
an electric marker writing tenths of seconds, J, Fig. 94, is pivoted 
to a post set on the hinged plate, H, and is adjusted by a screw so as 
to bring the writing point against the paper. It writes through an 
opening in the primary sheet upon the moving strip beneath. In 
order to keep the primary sheet from blotting this line, it is held up 
from the moving strip by two small brass clips, T, Fig. 96, and the 
time-line passes between these clips. 

"The glass pen, V, Fig. 94, which is used for the time record, is a 
form of capillary pen. To prevent clogging and uneven flowing, 
the opening in the point is made fairly large and the flow of ink con- 
trolled by a regulating air chamber. The upper end of the glass 
tube is inserted in a rubber tube which allows the point to move freely, 
and the tube is connected with a tambour. The rubber head of this 
tambour can be raised or lowered by a screw, and the ink thus made 
to flow slower or faster. 

" Besides the speed of the reactor's movements, it is desirable that 
the variations in the pressure of the pencil against the paper should 
also be recorded. The arrangement for securing a record of the 
pressure is shown in Fig. 96. Under the paper upon which the re- 
actor writes is a small table C, set into an opening in the base. The 
opening in which this table is set is situated immediately below the 
opening, N, of the hinged plate, so that the table occupies all of the 
writing space. The table is capable of an upward and downward 
movement, for it is fixed to the two bars, D, D, which are in turn 
fixed to the axis working in the pivot joints, M, M. The radius of 
movement of the table is, accordingly, the length of the bars D, D, or 
17 cm., and the direction of movement during a slight displacement 
is practically in a vertical line, The extent of movement of the table 


is magnified five times by means of the lever, F, which has its fulcrum 
at P. A disk on the outer end of this lever is in contact with the 
rubber of the tambour K, Fig. 94. The inner end of the lever, which 
is rounded, bears up against the table, making a sliding contact. In 
order to lessen the weight and consequent inertia of these parts, the 
table and its connections are made of aluminum. The long arm of 
the lever nearly balances the weight of the short arm together with 
the table and its supporting bars. The slight residue is counter- 
balanced by a light spring, L. This can be adjusted so that it will 
bring the table quickly back to position, but will not prevent a deli- 
cate response of the lever to a very light pressure on the table. The 
spring, L, as well as the tambour, with which a disk on the end of the 
lever is in light contact, are supported by a rod, shown in Fig. 94, 
fastened to the main base. The tambour is adjustable so that its 
head will just touch the lever when the table is in position. This 
apparatus responds with delicacy sufficient to easily record all the 
ordinary changes in pressure during writing. Tests with weights 
show that it will record changes in pressure of from 20 to 300 grams. 
"The remainder of the apparatus for recording pressure is shown 
in Fig. 94. The receiving tambour, K, is connected with the record- 
ing tambour 7, which writes on a long strip of smoked paper, E. 
This strip travels over the drum, D, and another drum 3.5 meters 
away. The drum D is clamped by an adjustable screw to the same 
shaft as the drum C, so that both of them can be driven together, or 
either one can be run separately by loosening the screw which clamps 
D to the shaft. Above the tambour pointer is a fixed pointer which 
traces a straight line with which to compare the pressure curve. 
The pressure curve is correlated with the speed curve on the moving 
strip, B, by means of one pointer of the double marker, L, which is 
in circuit with the marker J, on strip B." 

For a full account of records obtained from this appa- 
ratus, and methods of measuring the same, Mr. Freeman's 
article should be consulted. 


The following table shows in sigmas the time of con- 
tact of the various fingers with the key, and also the length 


of the interval between each of the contacts. The quan- 
tities in the table are averages from twenty-five cases. 

Finger I Interval Finger Interval Finger Interval Finger 

(Index) after I II after II III after III IV 

Avg. M.V. Avg. M.V. Avg. M.V. Avg. M.V. Avg. M.V. Avg. M.V. Avg. M.V. 

85 1.0 37 5.3 80 5.0 40 6.2 100 2.2 45 7.0 100 1.8 


75 1.2 116 9.8 82 1.1 89 1.5 127 3.1 51 2.3 99 2.0 

The following table shows the average length of line 
drawn in twenty trials in successive tenths of a second. 

1st 1/10 2nd 1/10 3rd 1/10 4th 1/10 5th 1/10 

Avg. M.V. Avg. M.V. Avg. M.V. Avg. M. V. Avg. M. V. 


2.1 1.3 17.2 4.1 43.5 6.7 37.2 7.9 12.2 5.1 


1.7 1.2 11.3 6.2 31.1 7.9 36.4 8.1 9.4 3.2 


The experiments here suggested are referred to again 
in later exercises, especially in Exercise XIX. The ap- 
paratus is not different from that required in general for 
recording movements and measuring their characteristics. 



The apparatus commonly used for measuring reaction 
time is, first, a timepiece that will measure very short in- 
tervals of time, and second, the accessories which make 
it possible to measure by means of this timepiece the in- 
terval which elapses between the signal to react and the 
reaction movements. 

The simplest arrangement for measuring reactions re- 
quires no other apparatus than that which has already 
been described. Let a contact key, such as that used in 
Exercise XIV, be connected with an electric marker which 
records on a kymograph. The marker should be allowed 
to trace in parallel with the record of a tuning-fork. In 
this case the tuning-fork record should be made if possible 
from a fork vibrating at the rate of 500 times a second. 
Another marker should be introduced to record the mo- 
ment at which the stimulation for the reaction is given to 
the reactor. This marker should be connected with a 
second key, which makes a sound that serves as the signal 
to react and at the same time completes the electric cir- 
cuit passing through the marker. With these three 
records that of the reaction key, that made by the tuning- 
fork, and that made by the apparatus which gives the signal 
for reaction it is possible to determine the number of 
tuning-fork vibrations which lie between stimulus and 
reaction. Such determinations as would be possible with 
the arrangement just described involve a great deal of 
counting of tuning-fork vibrations; consequently, other 



forms of apparatus have been devised and are in common 
use, which obviate the necessity of such elaborate counting. 
A simple and very convenient form of chronoscope 
which can be used in connection with a tuning-fork to 
obviate counting is the Ewald Chronoscope, which is rep- 
resented diagrammatically in Fig. 97. Its essential parts 
are a pair of small electromagnets M which can be put in 
the circuit of an electric current which is made and broken 
by a 100-vibration fork. 
This pair of electromag- 
nets, being supplied with 
an electric current every 
1-100 of a second, can 
be made to set in oscilla- 
tion a little plate P which 
hangs in front of the 
magnets and is drawn 
away from them by the 
spring S whenever the 
current is broken. The 
plate thus set in vibration 
is connected by means of 
a short rod with a fine 
toothed wheel which has 
the form of an ordinary 

clock-work ratchet. Each time the rod is thrown for- 
ward into this wheel it advances the wheel one tooth, 
so that the rate of the movement of the wheel will be 
one tooth in every 1-100 of a second. Connected with 
the ratchet wheel is a hand T, which moves across the 
face of a circular dial, graduated so that one graduation is 
equal to each tooth of the ratchet wheel . If no w the electric 
current from the tuning-fork is allowed to pass through the 
electromagnets for any length of time, the hand on the dial 

FIG. 97 


of the chronoscope will travel over a number of gradua- 
tions equal to the number of vibrations of the fork which 
occur between the beginning and the end of the introduc- 
tion of the current into the electromagnets. In order to 
measure reaction times, connections should be provided 
such that the current from the fork will begin to pass 
through the electromagnets at the instant that the signal 
for the reaction is given, and will cease to pass through 
the electromagnets the instant that the reactor moves his 
hand. The proper connections are indicated in Fig. 98. 
The current from the battery B passes at F through the 
fork and at K through a key. This key is a break key; 

FIG. 98 

that is, when in position, the current passes through it, 
when the experimenter presses down on it the current is 
interrupted. From the points / and Z, at which the wires 
from the battery and fork connect with the key K, a second 
circuit leads through the Ewald Chronoscope (7, and a 
second key M. The key M is the reactor's key and at the 
beginning of the experiment is closed. This second cir- 
cuit supplies the path for the current when the contact at 
K is broken. If the circuit is broken by taking the finger 
off the key M, as it is when the reactor raises his hand, 
the current will no longer pass through the chronoscope. 
A reaction time experiment in its simplest form can be 



carried out with this apparatus by requiring the reactor 
to prepare for the experiment by pressing down upon the 
key M. The making of this contact without a breaking 

FIG. 99 

From Wundt's "Grundziige der physiologischen Psychologie " 
5th Ed., Vol. Ill, p. 339 

of the current at K is not sufficient to drive the current 
through the Ewald Chronoscope, because of its relatively 
high resistance. The reactor may, accordingly, make 
preparation in this way for the experiment without starting 


the chronoscope. The signal to react is given by the ex- 
perimenter by striking the key K. The simplest signal for 
reaction is the sound thus produced at the key K. The 
Ewald Chronoscope will be set in motion by the current 
coming from the electric tuning-fork at the instant that 

Fio. 100 

From Wundt's "Grundziige der physiologischen Psychologic," 
5th Ed.. Vol. Ill, p. 392 

K is depressed, and it will continue in operation until the 
hand of the reactor is lifted from the key M. The number 
of 1-100's of a second that the chronoscope was in operation 
can now be read on the scale and this is the reaction time. 
A more elaborate chronoscope is that which is known 


as the Hipp Chronoscope. This chronoscope consists of 
three essential parts. First, there is a clock-work which 
is represented in outline in Fig. 99. Instead of being 
controlled by a pendulum as is an ordinary clock, the works 
of the Hipp Chronoscope are controlled in their rate by 
the vibration of a fine metallic rod, F, Fig. 99. This rod 
vibrates at the rate of 250 times per second, and releases 
the clock-work at a high rate of speed, and at the same time 
regulates that speed so that the clock-work moves uniformly 
under its control. The same works which are seen from 
in front in Fig. 99, are shown from the side in the middle 
section of Fig. 100. The clock-work drives the toothed 
wheel K 1 (Fig. 100) but not the similar wheel K 2 . 

The recording pointer and dial constitute the second 
part of the chronoscope. The dial is made up of two scales, 
the upper one of which indicates when the pointer moves 
over a single graduation one one-thousandth of a second, 
or a sigma as it is called. The lower dial and pointer 
which are connected with the upper by proper reducing 
gears, record tenths of a second. When the shaft of the 
recording parts is thrown into the clock-work so as to 
move at the same rate as the clock-work, as it is when the 
stylus h is held in the wheel K lf the period of time that 
elapses between the beginning of the movement of the re- 
cording apparatus and the end of that movement can be 
read on the dial. 

The third part of the Hipp Chronoscope consists of a 
pair of electromagnets E lt E 2 (Fig. 100) which make it 
possible to control the reading part of the apparatus so 
that it will move with the clock-work or come to a stand- 
still. Between the electromagnets is placed a metallic 
plate ra, which is also controlled in some cases by means of 
springs not shown in the figure. This metallic plate can 
be drawn up or down by the magnets, according as the one 


or the other carries an electric current. Whenever TO 
moves downward it acts through H 3 so as to throw h into 
the wheel K v When m moves upward it throws h into 
the wheel K 2 and prevents the hands of the dials from 
moving. The accessory connections are very similar to 
those shown in Fig. 98, and are shown in full in Fig. 101. 
Two keys, K R and K E are placed in circuit with the lower 
magnet of the chronoscope. For the sake of simplicity let 
it be assumed that the plate m is in this case held away 
from the lower magnet by a spring. If now either key 
is opened no current will pass through the chronoscope. 
The clock-work may be set in motion without carrying 
the recording arms on the dials with it. Preparation for 

Fio. 101. 

the experiment may be made by requiring the reactor to 
press down the key K R . By pressing down upon this 
key, the reactor closes one of the breaks in the electric 
circuit, but there will still be a break at the key K E , and 
the current will, therefore, not pass through the electro- 
magnet in the chronoscope. After the reactor has pre- 
pared for the experiment by closing the circuit at K R , the 
experimenter may start the clock face by pressing upon 
key K E . This pressure upon key K E allows the current 
from the battery B to pass through the electromagnets in 
the chronoscope (L) and sets the recording arms on the 
dial in operation. The movement of the recording face 
will continue until the reactor again breaks the circuit by 


lifting his finger from K R . The reading on the dial will, 
therefore, correspond exactly to the interval that elapses 
between the pressure upon K E and the breaking of the cir- 
cuit at K R . X is a commutator which changes the di- 
rection of the current after each reaction. 

The experiment may be made more elaborate by intro- 
ducing other pieces of apparatus into the circuit. Thus, 
instead of allowing K E to close the circuit through the clock, 
the experimenter may arrange a secondary circuit whereby 
his pressure upon K E will cause an electric hammer to sound. 
The electric hammer-head can be made to close the circuit 
through the clock, other connections remaining the same 

FIG. 102 
From the catalogue of Zimmermann, Leipzig, Germany 

as in the first case described. Fig. 102 represents an elec- 
tric sound hammer. The sound hammer consists of a 
long handle P and a metallic hammer-head H. An elec- 
tromagnet E is so placed that it may draw the rod P down- 
ward and thus cause a sound by the forcible contact of 
H upon the metal plate under it. Electric connections 
for sounding the hammer are made through the magnet. 
The necessary contacts for starting the clock may be made 
by connecting one wire with H through the rod P and a 
second wire with the plate upon which H strikes. The 
current will now pass through the two wires when H and 
the plate under it are in direct contact. A convenient 


substitute for a sound hammer is the common telegraph 
sounder. This is essentially a sound hammer in princi- 
ple and has adjustments for controlling the extent of the 
movement and the consequent intensity of the sound. 

In practical use the Hipp Chronoscope requires very 
careful handling. In the first place, care must be taken 
to keep the magnets from becoming permanently magne- 
tized. If the current is allowed to pass through the mag- 
nets in the same direction a number of times in succession 

Fio. 103 
From the catalogue of Zimmermann, Leipzig, Germany 

there is danger that they will gradually become so strong 
that the release under the tension of the spring when the 
current is broken will be slow, and an error will thus be in- 
troduced into the readings of the chronoscope. For this 
reason the current should be changed in direction each 
time it is sent through the electromagnets of the chrono- 
scope. This can be readily done by means of a simple 
mercury commutator. 

In the second place, the relation between the strength 
of the current and the tension of the springs should be so 


adjusted that the chronoscope acts uniformly. The best 
method of establishing and maintaining this relation is to 
adjust and test the chronoscope by comparing it with a 
mechanical device which will operate the clock during 
the time which elapses between the beginning and the end 
of the movement of a heavy falling body. One of the 
most common forms of control apparatus is the control 
hammer, represented in Fig. 103. This makes and breaks 
two electric contacts at A and B in the course of its fall. 
The hammer H is held in position by the magnet M until 
the clock is started, when it is released. As O passes A in 
the descent of the hammer the clock is started. When 
H strikes B the clock is stopped. The chronoscope should 
not vary in successive trials with such a mechanical de- 
vice, if it does it needs readjustment. Other forms of 
control apparatus may be made to operate contacts, 
or a freely falling body may be used. 

There are a number of pendulum chronoscopes which 
have been devised as substitutes for the various clocks 
and electrical apparatus described. The range of use- 
fulness of these pendulum chronoscopes is by no means 
as large as that of the various types of apparatus des- 
cribed. Reference may be made, if one desires to be- 
come familiar with this type of apparatus, to descrip- 
tions by Professor Scripture in his New Psychology, 
Chapter IX, page 155, and Professor Sanford in the 
American Journal of Psychology, Vol. XII (1901), pp. 

The procedure in reaction experiments is complicated 
by the necessity of securing the maximum attention of the 
reactor just before the signal is given. It has been found 
advantageous to warn him two seconds before the stimulus 
in order that he may be fully prepared. If the warning 
is given less than two seconds in advance, the period of 


preparation has been found to be in general inadequate. 
If it is given more than two seconds in advance, attention 
flags before the arrival of the signal to react. 

It is advantageous, where this is possible, to remove 
the reactor far enough from the apparatus so that he will 
not be distracted by preparations. 

In all of the earlier investigations of reactions the equip- 
ment which has been described up to this point was re- 
garded as entirely adequate. It was assumed that the 
hand reaction under the simple conditions presented was 
uniform. It has been made very clear by recent investi- 
gations that this is not the case, and that there is much 
productive information to be gained from an examina- 
tion of the form of hand movement involved in simple re- 
actions. In the simplest case this investigation of the 
form of movement may be altogether dissociated from the 
measurements of duration. The finger of a reactor may 
be placed on a receiving tambour or on a lever which is so 
placed that it records directly on a kymograph, and the 
reactor may be required to react to a signal. If the record 
thus secured is accompanied by a standard time-line, the 
time of the various phases of the movement may be 
measured by counting the vibrations in the time-line. 

If an electrically controlled chronoscope is at hand, a 
graphic record may be taken from a tambour, and at the 
same time the chronoscope may be used to measure the 
reaction time. In this case the tambour should be sup- 
plied with a metal sheet and the reactor should have a 
metallic cap fixed to his finger. Fine wires should be led 
from the metal plate of the tambour and finger cap to the 
chronoscope and the battery that supplies the current 
for the chronoscope magnet, just as in Fig. 101 connec- 
tions passed through the clock from the reactor's key, K R . 
The reactor should bring the two metal plates into contact 


in such a way as to depress the rubber of the tambour. 
If this tambour is connected with a recording tambour, 
any change in the pressure of the finger will be recorded 
by the recording tambour. The reaction from this ap- 
paratus will proceed exactly as in the earlier cases by the 
lifting of the reactor's finger. Some care must be exer- 
cised to make the tambour sufficiently large so that the 
rate of movement of the rubber surface in its recovery will 
not be equal to, or greater than, the rate of the finger as it 
is raised in reaction. If the movement of the rubber is 
equal to, or greater than, that of the finger, the plates of 

FIG. 104 

the tambour and finger cap will be kept in contact even 
after the reactor begins to raise the finger. The reaction 
time will, by this purely mechanical process be somewhat 
exaggerated in length. This difficulty may be overcome 
by making the rubber face of the tambour relatively large, 
when the rate of its movement will be slow. 

An especially devised apparatus, which gives great 
range to this type of experiment, is shown in Fig. 104. 
On a heavy table H is erected a firm post G. This post 
should be about 15 cm. high and has screwed into its top 
a long strip of spring brass seen from the side in the figure 
at A. This strip of spring brass is 25 cm. long and 5 cm. 


wide. The spring when in use is depressed with its at- 
tached parts to the position indicated by the dotted lines. 
The dimension of the spring may be determined empiri- 
cally. It must be made of such size that its rate of oscilla- 
tion, when set free from its depressed position, is slower 
than the rate at which the hand or finger of a reactor is 
lifted in making a rapid reaction movement. The method 
of determining the proper rate of the spring is very simple. 
Several reactors are required to record on a kymograph 
by means of a simple lever the rate of their movements 
when lifting the hand as they would in an ordinary reac- 
tion. The spring is then made enough slower than the 
slowest of these hand movements to insure its rising more 
slowly than the hand of any reactor. The rate of the 
spring is, on the other hand, fast enough so that any 
gradual movement of the hand upward will not separate 
the finger and spring. Put in other terms, the spring will 
follow faithfully any slow upward movements of the hand, 
and it can, of course, be pressed downward by any down- 
ward movement of any rate whatsoever. 

At the end of the spring A there is attached at a ful- 
crum C the reaction key M. This key is closed by press- 
ing downward at M . The contacts n n, n'n' are brought 
together by such a downward pressure. There is a small 
spring at B against which the downward pressure applied 
by the finger at M is exerted. This small spring B tends 
to separate the contacts n n, n'n'. If A were a rigid rod 
instead of a spring, the small spring B would be brought 
into action at the slightest movement of the finger upward 
from M. But since A is a heavy slow spring and B is a 
small rapid spring, we have the following complex results of 
finger movements at M. When the finger is pressed 
downward at M it overcomes the spring B and brings to- 
gether the contact n n, n'n'. At the same time it flexes the 


spring A for a short distance. As soon as n n, n' 
firmly closed, any further downward pressure will 
pended altogether in the flexion of A. In practical use 
the spring A is flexed for some distance after the contact \ 
are firmly closed. This flexion of A beyond the point oi' 
closing the contacts may be described as the surplus 
flexion of A. 

If now the finger rises at a rate which is slower than 
that at which the large spring A would naturally recover 
its position of rest, the contacts n n, n'n' will remain 
closed through the whole of what has been called the sur- 
plus flexion of A. If the finger is pressed downward at 
any rate whatsoever, the contacts will remain closed. 
There is one case of movement in which the contact will 
be immediately broken. That is the case of a rapid re- 
action movement upward. If the movement upward is 
more rapid than the rate of the spring A, as it is for ex- 
ample in a reaction movement of the ordinary type, 
then the lifting of the finger from M will immediately call 
into play the small rapid spring B, and n n, n'n' will be 
separated by B without reference to the slow upward 
movement of A. 

This combination of springs gives us all the conditions 
necessary for maintaining a contact at n n, n'n' until the 
reaction takes place, while it leaves the hand free to move 
downward at any rate whatsoever, or to move upward 
at any rate slower than that of the spring A. 

The method of recording any upward or downward 
movements of the reacting hand is to attach a long lever 
to the post G at 0, and place one end in contact with the 
key M while the other end traces on a kymograph at E. 
This connection may be of the form shown in the figure 
or it may consist in some other form of lever connection. 
A marker may be attached to the lever as at F and 


may record the signal to react and, if desired, the reaction 

rf The extent of the hand movement to be recorded and 
'0. e necessity of trying long series of experiments in rapid 
1 accession make it convenient to use on the kymograph 
che long belts of paper described under Exercise XL The 
adjustment of a belt running in the vertical differs from 
the adjustment of a horizontal belt somewhat, but will be 
easily mastered by one who has learned to adjust horizontal 


Table of reactions expressed in one-hundredths of a 



No. of deter- 


No. of deter- 


Avg. M.V. 

Avg. M.V. 

19.5 4.3 
15.5 1.5 
22.6 5.9 


17.4 1.7 
15.0 2.0 
14.3 5.0 



For complete statement as to forms of reaction, see 
Yale Psychological Sludies, New Series, Vol. I, pages 


If the reaction is to be to visual stimuli a very simple 
exposure apparatus may be set up as follows. Connect 
with the key which is to make the contact a long lever. 
Let the end of this lever be hidden from the reactor behind 
a screen when the key is not in use. Behind this screen 
various colors, letters, or other visual stimuli may be 
attached to the lever. The screen should be of such a 
size that when the key is closed the end of the lever 


with its attached visual object will appear through an 
opening in the screen, and this become visible to the 
reactor. A simple form of shutter on this principle is 
shown in Fig. 105. In this figure the opening in the 
screen and the key are shown. The reactor is placed on 
the opposite side of the screen. 

A simple visual exposure apparatus consists of a board 
which can fall between two guides and which has in 
its center a hole through which a series of letters or 
figures can be exposed. In its first position the board 
covers these letters, in its last it exposes them to view and 

FIG. 105 

at the same time by means of metallic strips at the side 
makes any desired electric contacts. 

The disadvantages with the fall apparatus and also 
with the simple shutter shown above are: first, both 
pieces produce a sound as well as expose letters to view; 
and, furthermore, since there is a general movement of 
the whole field during the exposure there is large probabil- 
ity that the observer will be distracted by the general 
movement in the field. An ideal visual exposure apparatus 
is one which presents the visual object to view without 
any apparent motion in the field. In order to obtain these 
ideal conditions the light which falls upon the visual field 


should be cut off and turned on, rather than the field itself 
either moved or exposed by a falling shutter. 

It is very easy to control the light which is to fall upon 
a given visual field provided this light is brought to a focus 
as indicated in Fig. 106 A. By means of the small bright 
light L, the lenses / and A 7 , and the diaphragms S 1 , S 2 , 
and S 3 , the field O is easily illuminated or darkened by 
a shutter placed at H. The visual field is enclosed in 
a dark box and is so placed that when the rays of light are 
cut off by the shutter H the field is entirely invisible be- 

N M 


Fio. 106 

cause it stands in the dark. When the shutter opens, 
the field is exposed and its whole area will be visible at 
the same time. There will be no movement anywhere 
visible in the field. The exposure shutter may in this case 
be of a very simple type, consisting of a metallic wheel 
shown in detail in Fig. 106 B. At its outer part this disk 
is made adjustable so that openings of different sizes may 
be made at T. The shutter is made to rotate by attaching 
a weight W to its axis. The shutter is stopped by means 
of a brake, which consists of a wedge (R) attached to the 
shutter and a felt pad D which is held in position by a 


spring not shown in the figure. This break holds the 
shutter firmly in position after it has exposed the visual 

.Fig. 107 shows a key designed by Professor Scripture 
to give tactual stimulations. A heavy rod H is set into 
a handle and is connected with a wire. The flexible rod 
N is placed above H and in electrical contact with it. 
At the end of N is a small hard rubber point with which 
pressure is to be exerted on the skin. When this point 
is brought against the skin, it breaks the contact between 
N and H. 

It is frequently desirable to make a number of different 
contacts simultaneously in order to set different pieces of 

FIG. 107 
From the " Studies from the Yale Psychological Laboratory," Vol. III. 

apparatus in action at the same instant. A very good 
means of securing simultaneous contact has been devised 
by Professor Angell. It consists of a large wooden cyl- 
inder upon which are fastened at certain points metallic 
strips which extend over only a part of the circumference 
of the cylinder. Adjustable brushes are set against the 
surface of this cylinder. When in contact with the wooden 
cylinder these brushes make no electric contact. If now 
the cylinder is turned until the brushes come in contact 
with the metallic strips referred to above as fastened to 
the surface of the cylinder, a contact will be made between 
each brush and the metallic strip with which it comes into 
contact. By carefully adjusting the brushes so that two 


or three come into contact with their respective metallic 
strips at the same moment, a series of contacts can be 
made simultaneously. The advantage of the large cyl- 
inder is that the contacts can be adjusted to secure abso- 
lute precision in the making or breaking of the contacts. 

The supplementary experiments which require records 
of hand movements are closely related to the experiments 
required in Exercise XVII. The apparatus there described 
in full may be used for these experiments. 



Most of the apparatus necessary for this exercise has 
been described. For choice reactions the colors or sounds 
may be produced by means of one of the exposure screens 
or by means of two or more hammers placed in various 

FIG. 108 

From Wundt's " Grundziige der physiologischen Psychologie," 
5th Ed., Vol. Ill, p. 403 

positions. Reaction keys for more than one ringer can 
be constructed by bringing together in compact form a 
series of finger pieces which control electric contacts. 

For recording the reactions of articulation the simplest 
device is to require the reactor to move his hand at the 



same time that he utters the word. This method is not 
satisfactory because some error arises when the reactor 
attempts to make two simultaneous movements. The 
only satisfactory method is to secure a record from the 
articulation movement itself. For this purpose a special 
voice key may be used. The most satisfactory form of 
voice key is represented in Fig. 108. A large mica plate 
is placed at the end of a funnel. The mouth is placed 
against the opening of this funnel M and a word is sounded 
into the air chamber. The vibrations, as well as the change 
in the pressure of the air, set the mica plate in oscillation. 
On the surface of this mica plate is a small platinum con- 
tact C. An adjustable point controlled by means of the 
screw is set so that it is in contact with the platinum at- 
tached to the mica plate. Wires pass from the two poles 
of this contact. These two wires, instead of being con- 
nected directly with the chronoscope, are first carried to a 
relay apparatus. The relay is a piece of apparatus which 
may be set so that when it is once released by means of an 
instantaneous electric current, it will not return to its 
original position unless it is deliberately set in position. 
A simple form of relay is represented in Fig. 108 at 
the right. A metallic strip T is connected with the spring 
F which tends to draw it away from the electro-mag- 
nets. When the apparatus is set ready for use, the 
metallic strip T is held in position by the magnets which 
are supplied with a current through the mouth-key de- 
scribed above. If now the current is broken even an in- 
stant by vibrations in the voice key, the metallic strip T 
is released by the magnets and responds to the tension of 
the spring F. When once it is. drawn away from the 
magnets it does not return with the reestablishment of 
the current in the magnets. At the lower end of T are 
electric contacts which are broken or made with the move- 


merits of the strip. In this way an intermittent make and 
break, such as that which results from the vibration of the 
voice key above described, is converted into a permanent 
break the moment the current in the voice key is inter- 
rupted. From this point the connections from the chrono- 
scope are as from the regular break key and require no 
special description. 


The following discrimination times are reported for 
reactor B, for whom simple reaction times were reported 
on page 196. These times are long. They are expressed 
in one-hundredths of a second. 

Avg. M.V. 

10 Discriminations of Green and Yellow 44.7 3.6 

10 Discriminations of Red and Blue . .52.8 13.8 

Articulation times from reactors who pronounced 
printed words exposed to view are given in sigmas as fol- 
lows: Train, 560; boat, 570; blot, 440; tank, 1100; dark, 

Association times for free associations are as follows: 
Boat water, 980; train smoke, 850; bank money, 1270. 


A very good discrimination experiment consists in pre- 
senting to the reactor a number of different pairs of colors 
which differ qualitatively from each other in various de- 
grees. Thus, one pair should be made up of a certain 
red to be designated as R and a second red slightly differ- 
ent from the first and designated as R'. A second pair 
of colors should be made up by using R and a quality of 


red which differs from it more than R'. This third red 
will be designated as R". In this way a series of reds 
can be made up, all of them differing from the original 
red R but in increasing degrees. R and R' should now 
be shown to the reactor with R sometimes on the right- 
hand side and sometimes on the left, and he should be re- 
quired to react in every case with the hand corresponding 
to R. It will be found that the time required for discrimi- 
nation decreases as the difference between the two qualities 

Various types of associations between words may be 
measured. The time for so-called free association is the 
shortest of the association times. If a word is presented 
and the reactor is required to associate some other word 
with it but no restrictions whatsoever are placed upon the 
direction of this association, we shall have such pairs as 
chair desk, chair floor, chair table. A variety of re- 
strictions may be imposed upon the reactor. Thus, he 
may be required to name a second object belonging to the 
same class as the first object or he may be required to men- 
tion a part of the object named. He may be required to 
name the general classes to which the object belongs. 
These restrictive associations require much longer time 
than the free associations and they vary a great deal as 
compared with each other. 

The experiment of striking out a's may be made with 
any page of printed matter that is conveniently at hand. 
For purposes of comparison in a series of experiments, it 
is desirable that the distance over which the hand must 
travel in marking out a given number of a's shall be uni- 
form. For this reason it is better to prepare tables which 
are composed of letters of the alphabet irregularly arranged 
and containing a standard number of each one of the letters. 

A convenient method of preparing such tables was sug- 


gested by Dr. Whipple. Let 100 printer's types of each 
of the letters to be used be "pied" and then set up as a 
regular press form. The form may need a little editing 
to prevent letters from being repeated in close proximity 
to each other, but in general the chance order of the letters 
can be accepted without further change. 



For the measurements of the duration of writing ac- 
tivities the apparatus and method described on pages 176- 
180 are to be used. A very simple form of this apparatus 
can be set up by merely writing over a moving strip of 
paper through a sheet of carbon paper. The moving paper 


I I 
7 8 

Fio. 109 

I I I I 
12 13 4 15 

which receives the record can be drawn along as suggested 
on page 175 by a kymograph, its rate being recorded by a 
time marker. Figures 109 and 110 show fluctuations in 
the rates of movement during the writing of different parts 
of the letters a and &.* 

A recorder which will very conveniently show the move- 

* These two figures have been supplied by Mr. F. N. Freeman from a general 
study in which he is engaged. The letters a and b in the figures are marked off 
in sections which correspond to distances of 1 mm. in the original written letter. 



ments of different parts of the hand during writing is rep- 
resented in Fig. 111. It consists of a spring A which fits 
closely on the hand and carries the rod BB. This rod 
runs forward far enough so that the distance of its end 
from the wrist and elbow is the same as the distance from 
the wrist and elbow to the pen held between the thumb 
and fingers. This equality in length insures a record of 
the hand and arm movement which is on the same scale as 
the writing done by the pen, and thus comparison is made 




12 13 14 15 16 17 
FlG. 110 

21 22 23 24 25 26 

direct and easy. At the end of the rod B is placed a metal 
or glass tube C, and through this falls a tracing point DD 
of the same form as that which has been described under 
Exercise XIII, page 170. The record is made by allowing 
the reactor to write naturally on the paper with the pen 
held between the thumb and fingers; the tracer in the 
meantime making a record of all of the movements which 
the hand makes during the writing. The written letters 
are the records of the finger movements plus all of the hand 


and arm movements, except such hand and arm movements 
as are made during the intervals between the words, when, 
of course, the writing pen is raised from the paper. The 
tracer is carried on the paper by its own weight and re- 
cords all that the hand and arm do, but gives nothing of 
the pure finger movement. By a comparison of the 
written letters with the tracer record it is easily possible to 
determine what part of the whole work is done by the hand 
and arm and what is done by the fingers. Two records 
of writing with their corresponding tracer records are 

shown in Figs. 112 
and 113. 

In order to se- 
cure the moving 
point which is re- 
quired in the third 
part of the experi- 
ment, the follow- 
ing combination of 
apparatus already 
described may be 
arranged. A belt of 
paper is carried be- 
tween two kymograph drums. On this belt of paper a 
line is drawn obliquely across the paper. The belt is set 
in rotation by means of the drum, and the reactor is allowed 
to look at it through a narrow slit in a large shield, the slit 
extending in a direction perpendicular to the direction of 
movement in the paper. The whole is represented in Fig. 
114, where A B is the belt and S the shield. The shield is 
intended to be large enough so that the observer sees 
nothing of the belt beyond it. The shield thus prevents 
the reactor from seeing the line on the rotating belt, but 
gives him the opportunity of seeing the single part of the 

Fio. Ill 


line which lies directly under the opening in the shield. As 
the belt moves forward, the part of the line which the re- 
actor can thus see will gradually move upward or down ward 

Vutf/V?/ //.B 4 - 

\JA \A!A/ ^--'' ^. 

i i 

10 n 


FIG. 112 

along the line of the opening in the shield. If the oblique 
line drawn across the belt is a regular line, and the drum 
moves at a uniform rate, the point seen by the reactor will 
travel up and down the slit at a uniform rate in a given di- 
rection. Irregularities in the line will result in irregular 
movements of the points seen through the slit. If, therefore, 

FIG. 113 

it is desired that the reactor strike at a point which is mov- 
ing uniformly, a single line is drawn across the moving belt. 
If it is desired that a certain change in the rate of move- 


ment of the point shall be introduced, irregularities of the 
type indicated in Fig. 115 may be drawn in the long line on 
the belt of paper. In order that the effort to react may be 

FIG. 114 

properly timed with reference to the irregularities in the 
line, a sounder may be set up and the signal to react may 
be given by the experimenter by means of this sounder. 

The movement of the hand in aiming to strike this mov- 
ing point can be studied in several ways. First, the shield 

Fio. 115 

which covers the traveling line may be made of paper and 
a carbon sheet may be placed under it. The hand may 
move across the sheet of paper, tracing a pencil line from 
the point of starting until it reaches the point seen through 
the slit. All irregularities in the movement will thus be 

Fio. 116 

recorded on the same paper as the line which supplied 
the point. Such a record is shown in Fig. 116 in the line 
abcdE. The reactor in this case was directed to move his 


pencil from the position a where it was held just at the 
bottom of the slit until he could strike the point traveling 
downward. It will be noticed that he moves rapidly up- 
ward at first, then pauses and finally makes his last move- 
ment with increasing rapidity to the end. 

Second, a string may be connected with the hand in such 
a way that all forward movements of the hand shall draw 
out the string and indirectly make a record. This method 
does not permit movements in different planes to be re- 
corded without complicating the apparatus indefinitely. 
Movement in one plane can, however, be conveniently 
recorded by such a string. The string should be wound 
around a cylinder. This cylinder is supplied with an inner 
coil spring which keeps the string wound up under small 
tension. It is also fitted into a system of cogs so that any 
turning of the cylinder as a result of pulling the string 
results in the movement of a system of wheels and ulti- 
mately of a pointer. This pointer may be allowed to 
trace upon a smoked paper, thus giving a record of the 
amount of the hand movement. Since such an arrange- 
ment as this records the hand movement only in one 
direction, the conditions of the experiment should be so 
arranged that the reactor will move his hand chiefly in 
one plane. One of the best directions in which to allow 
the string to unwind is downward from directly above 
the slot in which the moving point appears. 


A device similar to that described above for recording 
the movement of the hand may be attached to any part 
of the arm during writing, in order to record the behavior 
of this part of the arm while the fingers are forming the 


A method which has been employed for the determina- 
tion of the pressure of the different fingers during writing 
consists in mounting on the pen a number of small tam- 
bours. One is placed under the thumb, one under the 
first finger, and one under the middle finger. From these 
various tambours connections are made with recording 
tambours and a triple record is taken on a smoked surface 
showing the various changes in the pressure of each of 
the fingers against the writing pen. 

The pressure exerted by a person writing on a sheet 
of paper can be readily measured by means of the appa- 
ratus described on pages 179-180. 


No special apparatus is required for Exercise XVIII. 
It is desirable that the figures to be used in this exercise 
should have a sufficient degree of uniformity to make com- 
parison between the different figures possible. To this end, 
the figures should all be made up of the same constituent 

FIQ. 117 

lines, these being combined in various different arrange- 
ments but having the same absolute dimensions and posi- 
tions in space. 

A series of tests of this type is fully reported and dis- 
cussed in the Yale Psychological Studies, New Series, Vol. 
I, No. 2, pages 349-369. 




There are a number of methods of recording the rate 
of vibration of sung or spoken tones. The most familiar 
methods are those of the phonograph. In this apparatus 
a voice vibration is delivered against the mica or glass 
diaphragm. The diaphragm by means of a sharp stylus 
cuts into a wax cylinder, recording the rate of the vibration 
which is imposed upon the diaphragm by the voice. The 
difficulty of reading such a record as this prevents the use 
of the phonograph for ordinary experiments in the psycho- 
logical laboratory. The principle of the phonograph may, 
however, by a slight modification be employed for the pur- 
pose of making records of voice reactions. 

Fig. 117 represents the apparatus employed by Dr. 
Cameron in recording the rate of voice vibrations and de- 
scribed by him in the Yale Psychological Studies, Vol. 1, 
No. 2, page 230, as follows: 

"The recording apparatus used in the present investigation con- 
sisted of a round rubber telephone receiver, a vertical cross section of 
which is represented in Fig. 117. The box (B) is provided with a 
cover (C) of the same material which may be screwed tightly to 
the face of the box. Between the from edge of the box and the 
cover is a diaphragm (D) of thin mica, which is held firmly in posi- 
tion by the cover, when screwed down. The diaphragm is 5 . 3 cm. 
in diameter. Glass diaphragms have also been used, but with 
less satisfactory results. 

"The cylindrical chamber (F) communicates directly with the air 
chamber back of the diaphragm. An aluminum mouthpiece (P) 
is attached to the outer edge of (F) by a small piece of rubber tubing. 
In the latter experiments a long flexible tube was substituted for this 
form of connection between (F) and the mouthpiece (P) . A small hole 
(R),3 mm. in diameter, is bored in the mouthpiece to allow the escape 
of the air forced into the chamber at the moment the tone is sung. 

"There is screwed to the front of the box a piece of brass (H), 
shaped as shown in the figure, and used for the purpose of holding 
the adjustable screws M and N. 


" M and N are held securely in position by the set screws, V and 
W. M and N are fitted with jewel bearings in which play the taper- 
ing ends of the steel axle A. 

"To the axle is attached the aluminum right-angle piece KL. 
K carries a straw (S) to the end of which is fastened the recording 
point X. This point is made of hammered brass, carefully cut to a 
point and polished. Such a point is fine enough to make a sharply 
defined line on smoked paper, and the lampblack does not adhere 
to it. 

"The other arm L is attached by a joint to a smaller link (O) of 
aluminum, which passes through an opening in the middle of the 
box cover and is fastened to the center of the diaphragm by a drop 
of glue. 

"There is thus provided a system of continuous levers from the 
outer surface of the diaphragm to the recording point, so that move- 
ments of the diaphragm caused by the singing of tones into the 
mouthpiece or by any other means are magnified and may be re- 
corded on a belt of smoked paper. Since it is desirable to obtain 
very long series of records, a long belt of smoked paper is used. The 
belt passes between two drums placed fifteen feet apart. It is smoked 
at one of the drums and after the record is made is shellaced from 

Such a record is reliable for determinations of pitch. 
It is not useful in determining the form of sound vibrations. 

A target can be very easily made of a drawing-board 
over which has been pasted a sheet of paper with concentric 
circles. A dart may be made up of a wooden shaft in the 
end of which has been inserted a sharp needle. The dart 
should be supplied with guides at the end opposite that 
carrying the needle. These can be made by cutting slots 
in the wooden shaft and inserting cardboard strips. The 
record may be taken by counting the score as in ordinary 
target practice. It is desirable also to record not only 
the distance from the center but also the quadrant in which 
the dart strikes the paper. 

The experiment which requires a stylus from which 
the reactor shall receive no sensation of pressure during 


writing may be carried on by means of the following de- 
vice. A pencil or small wooden rod the size of a pencil, 
is supplied at the end with a tube either of metal or glass, 
and through this tube is allowed to pass a writing stylus 
of the form described in Exercise XIII. The reactor will 
receive no pressure sensations from this pen, as the writing 
stylus is free to move upward and downward in the tube. 
The downward movement of the wooden rod will simply 
result in the sliding of the tube down the writing stylus. 
Experiments tried with such a stylus by a reactor who 
tries to write with his eyes closed indicate very clearly the 
results of withdrawing both visual and tactual sensations 
in writing activities and in the activities of drawing. 



The apparatus necessary for the first part of this experi- 
ment has been fully described in connection with Exercise 
XIV (see page 175). 

The second experiment requires no equipment other 
than a mirror and some patterns. It should be noted, per- 
haps, that simple geometrical patterns such as stars and 
rhomboids are better for the purposes of this exercise than 
more complex forms. 

A convenient method of preparing cards for the third 
part of this exercise is to secure plain cards and mark 
them with numerals, or paste on them kindergarten pict- 
ures or small pieces of colored paper which will be suffi- 
cient to distinguish them. 


The following table shows the changes in the length of 
contacts and intervals between contacts during practice 
with the unnatural or reverse order of finger movements. 
The first horizontal column shows the rate before practice. 
The second horizontal column shows the averages of ten 
cases after 200 trials. The third shows the average of ten 
cases after 400 trials, and the last shows the average of 
ten cases after 600 trials. 



At first . . . 
After 200. 
After 400 . 
After 600. 



























At first 






















After 200. . 

After 400 

After 600.. 

The following table shows improvement both in time 
and accuracy in following in five successive trials the same 
four-line pattern when the pattern was seen in a mirror. 

Corrective Movements 




Time in 




































The following table shows the time for 10 successive 
distributions of eighty cards which were marked with the 
numerals from 1 to 8. The spatial order of these cards 
in the first series of distributions was: 5-3-6-1-8-4-7-2. 

Distribution 1 2 3 4 5 6 7 8 9 10 

Time in seconds.. . .84 71 67 54 50 52 57 60 51 57 


When the order was changed to 6-4-3-7-5-2-1-8, the 
time of distribution for the first trial was 92 seconds with 
many tendencies to return to the first order. 


Experiments with writing seem to show that the time 
of writing for a given subject suffers little change from 
trial to trial. 

The simplest method of studying the control of the 
winking reflex is to hold before the eye a piece of glass 
and allow the experimenter to strike the glass directly in 
front of the eye. The degree of control will be shown by 
recording the number of times during which the observer 
is capable of resisting the reflex tendency to wink when 
the glass is struck, the glass being held in the successive 
trials at various distances. 

Methods of working with the typewriter will suggest 
themselves to any one who is unfamiliar with the manipu- 
lation of this machine. Simple copying of prose passages 
and a measurement of the time required for a given number 
of letters, together with the errors in these letters, serve 
very well to indicate the progress of the reactor who be- 
gins without previous training. The machine may be 
utilized for experiments even by experts if, instead of 
using the regular letters on the keyboard, certain colors 
or outline forms are substituted for the familiar keys. 
Each reaction will in this case be recorded as a letter, while 
the stimulus received by the reactor will not be a letter 
but the color or outline figure with which the key is cov- 



The apparatus required for recording taps has been de- 
scribed in connection with Exercise XIV, page 175. 

A metronome can be purchased at any establishment 
which sells musical instruments or from C. H. Stocking 


Table I shows some effects of distraction. The aver- 
ages are in sigmas. 

Average normal 

time for tapa 


without dis- 
traction, meas- 
ured by taking 

Fast met- 

Slow met- 

of other 


time of five 

successive taps 

Avg. M.V. 

Avg. M.V. 

Avg. M.V. 

Avg. M.V. 

Avg. M.V. 


183 6.3 

211 9.3 

218 7.5 

238 11.4 

193 7.7 


197 4.4 

209 8.3 

207 13.2 

254 10.1 

215 6.5 

The following results show the effects of fatigue. The 
first five taps in every 50 were counted and the total time 
was found as follows: 

910 sigmas, 920, 940, 900, 900, 880, 740, 970, 1,000, 1,240, 1,050, 
1,070, 1,080, 1,080, 1,300, 1,100, 1,095. 


Apparatus and methods for the first supplementary 

experiment are described, and reference to a complete ex- 



periment of this type is given under Exercise XVIII (see 
pages 214 and 215). 

The apparatus necessary for the exercise on muscular 
fatigue consists of a dynamometer such as already described 
under Exercise XIII. Such a dynamometer may be used 
continuously for a longer period of time, and a record 
taken of the amount of work done by the reactor. If 
the work is continued until the reactor is no longer able 
to move the dynamometer, the records will show the total 
amount of work necessary to produce complete fatigue. If 

Fia. 118 
From the catalogue of C. H. Stoelting Co., Chicago 

the work is not continued for a sufficient period of time 
to produce complete fatigue, the falling off in the records 
of the movements will show the increasing tendency 
towards fatigue. The disadvantages of using a dyna- 
mometer for such experiments is that it is extremely diffi- 
cult to secure a record of the total amount of work done. 
Other forms of instruments have accordingly been devised 
for recording and automatically measuring the amount of 
work a reactor can do. Mosso's Ergograph, which is one 
of the best known instruments, is represented in Fig. 118. 


This consists of a sliding bar connected with a weight at one 
end and a finger holder at the other end. The sliding bar 
is attached to a pointer which can trace upon a kymo- 
graph surface. The experiment consists in allowing the 
reactor to move his finger so as to lift the weight as long 
as it is possible for him to make any movement against 
the weight. The pointer records the form and extent of 
this movement upon the drum. An addition for the pur- 
pose of measuring the amount of movement can be made 
in the form of an automatic tape take-up which will draw 

up a tape each time 
the weight is lifted. 
During the test the 
hand and arm of the 
reactor are held rigidly 
in position, so as to 
eliminate as far as pos- 
sible the movements 
of all muscles except 
those which are being 
tested. The objection 
to such a weight ergo- 
graph is that at the 
end of a given experi- 
ment the finger can undoubtedly do some work, even 
though it can not do what it is required to do when 
the demand is made that it lift the large weight which is 
connected with the apparatus. The finger is in a con- 
dition in which it seems to be unable to do further work, 
not because it is completely exhausted but because it is 
too exhausted to do the large amount of work demanded. 
The remedy for this defect of the weight ergograph is to 
be found in substituting for the weight a spring. If now 
the reactor is allowed to pull against the spring instead of 


Fio. 119 

From the catalogue of C. H. Stocking Co., 


against a fixed weight, his early movements will indicate 
the absence of fatigue in the length of the movement. 
Later, the movements will grow smaller and smaller, and 
the amount of work demanded for each one of these small 
movements will, because the reactor is working against a 
spring, be reduced in quantity. The rate at which the 
reactor makes the movements should in all cases be a uni- 
form rate determined by a metronome which gives the 
signals for reaction. 

A simple ergograph supplied with the Harvard physio- 
logical apparatus and shown in Fig. 119 can be used in 
securing fatigue records from the index finger. 

More elaborate forms of ergographs have been devised. 
These elaborate forms are designed to secure greater isola- 
tion of the muscles exercised. It is possible in most of 
the simple ergographs for the finger which is called upon 
to do the work to receive much assistance from the other 
parts of the hand and arm which it is not intended to 
test. An elaborate device is described by Professor 
Bergstrome in the American Journal of Psychology, Vol. 
XIV, 1903, pp. 510-540. 



The best method of making up series of nonsense syl- 
lables consists in preparing a large number of such syllables 
on separate slips of paper. These syllables should be 
made up as follows: Start with the first consonant in 
the alphabet and the first vowel. Set down as the third 
letter of the syllable each of the succeeding consonants of 
the alphabet. Thus, we have bac, bad, baf, bag, etc. A 
second series should be prepared beginning with the 
second consonant of the alphabet as the first letter of the 
syllable. Add the first vowel, and in succession each of 
the remaining consonants of the alphabet. Thus we 
should have cab, cad, caf, cag, etc. Another series of 
variations can be produced by changing the vowel. Thus, 
the first combinations in the series would be bee, bic, boc, 
buc. In like fashion we should proceed with ceb, cib, cob, 
cub. After this series has been prepared, such combina- 
tions as "bad" and "bag" should be thrown out because 
they represent words. Such combinations as "caf" 
should also be stricken out because they resemble too 
closely common words of the language. After a series 
of combinations has been prepared by selecting non-sig- 
nificant forms, the whole collection should be placed in a 
box and thoroughly mixed. Any desired number may now 
be drawn from this thoroughly mixed collection and the 
character of the material will be sufficiently uniform for 
experimental purposes. The following list presents a few 
such syllables which may be used in some of the simpler 



exercises without the more elaborate preparation described 

Dap, Vac, Jaf, Lar, Bex, Bup, Nat, Hif, Lis, Mor, Cul, 
Zuc, Jut, Puj, Riv, Sal, Mik, Tex, Pij, Hoj, Dae, Cib, Rev, 
Bir, Huv, Vog, Gir, Jal, Kod, Dak, Pon, Sab, Def, Dor, 
Miv, Mel, Jub, Lig, Biz, Gar, Cem, Vub, Har, Dik, Rej, 
Ber, Lun, Hak, Jec, Fij, Ked, Tol, Soc, Gee. 

If experiments are to be performed with these syllables, 

FIG. 120 
From the catalogue of Z im merman n, Leipzig, Germany 

it is desirable that they should be quite legible. To this 
end they should be printed or copied on a typewriter or 
made up of gummed letters which have been pasted on 
cards. The cards may now be read off in series, each one 
being exposed to view by drawing off the one above it in 
a pack held in the hand. 

The rate at which the syllables are exposed is a matter of 
some importance. A device for securing greater uniform- 


ity of exposure than can be attained by simply handling 
the cards, is to paste the syllables on a kymograph and 

allow them to be seen 
through an opening in a 
screen as the kymograph 
is rotated with a slow, 
steady motion. This 
method has the disad- 
vantage of keeping the 
letters in constant motion 
during the whole period 
of exposure. A second 
device consists of a drum 
which is moved forward 

From the catatogue of Zimmermann, by means of a Weight and 

Leipzig, Germany , , 

a pendulum escapement. 

An electrical device for moving the drum forward at 
regular intervals is described by Professor Bergstrome in 
the American Journal of 
Psychology, 1907, pp. 206- 
238. A pendulum, instead 
of acting directly upon the 
drum, acts by means of an 
electric circuit which it 
makes with each oscillation. 
The apparatus may be com- 
plicated by the addition of 
a second pendulum so con- 
nected with the first that the 
intervals of movement in the 
drum shall be regulated dif- 
ferently during the exposure 
of the syllables and during the intervals between exposures. 

A very elaborate mechanism has been devised by 

FIG. 122 

From the catalogue of Zimmermann, 
Leipzig, Germany 


Professor Wirth, of Leipzig. It is shown in Figs. 120-121. 
In Fig. 120 A the apparatus is cased in and a single word 
appears as seen by the observer. In Fig. 120 B, the tape 
with syllables and the wheel on which this tape moves 
is shown. In Fig. 121 the weights and electrically con- 
trolled escapement for the apparatus are shown. When a 
current is passed through the electromagnets the syllables 
move forward. The rate of this movement is determined 
by the apparatus which sends the electric current into the 
magnets. This controlling apparatus is the pendulum 
shown in Fig. 122. By means of adjustable bobs the rate 
of this pendulum can be varied within wide limits. By 
means of contacts placed along the scale shown at 0, 
C, C, 0, in the figure the current can be sent as desired 
to the escapement apparatus. 


Table showing individual differences and differences 
in the same individual under varying conditions. 

Readings required to learn ten syllables. 





and seen 





















Table showing readings on first learning and after one 


first time 












Table showing difficulty of learning series of different 
lengths, all read silently. 


No. of syllables 







not lc>:iniod at 







The monochord required for memory of tones is de- 
scribed on page 111. 

For the other supplementary experiments, no special 
apparatus is required. 



Equivocal figures such as are required for the first part 
of the exercise are shown in Fig. 123, A and B. The 
record of the rate of fluctuation in these figures can be 
taken by means of a key such as that described on page 
175 and an electric marker such as that described on 
page 155. The observer is required to press down upon 

FIG. 123a 

the key when the figure appears in its first position and to 
lift the finger when the figure reverses. In some cases it 
may be necessary to record more than two types of experi- 
ence. In this case either two markers and two keys can 
be employed, or a code of signals can be worked out to be 
operated by the single key and marker. 



For the experiments on retinal rivalry a lens stereoscope 
can be used. (See pages 93 and 94.) 

A Masson disk is represented in Fig. 124. It consists 
of a white disk marked at intervals with black lines which 

Flo. 1236 

have all the same absolute width. The relative width of 
the successive lines as compared with the white area which 
is mixed with the black when the disk is rotated, grows 
constantly less from the center outward. The result is 
that the gray rings which appear when the disk is rotated, 

Fio. 124 

grow lighter and lighter from center to circumference of the 
disk. One ring will be found that is only intermittently 
visible. This is the ring for which a record should be 



The apparatus necessary for this exercise can be de- 
scribed best by exhibiting at the outset a complete form of 
the complication pendulum. Substitute devices of a sim- 
pler character can be made on the same general principle. 
Fig. 125 shows the complex form of complication pendu- 
lum described by Wundt. It consists in a pendulum 
PL the upper end of which is coupled by means of a cog- 
wheel with the pointer Z. Whenever the pendulum os- 
cillates, the pointer Z moves in front of a scale S. When 
the pendulum is in use, all of the mechanism behind the 
dial is hidden so that the observer can see nothing but 
the scale arid pointer Z as it moves backward and forward 
across the scale. By means of an adjustable connection, 
a bell can be sounded or an electric circuit closed when 
the pointer is at any desired point on the scale. The ex- 
periment consists in setting the bell so that it sounds when 
the lever is at a given point on the dial, and requiring the 
observer to determine the position of the pointer on the 
scale at the instant when the bell sounds. The rate of 
the pointer's movement can be modified by shortening 
or lengthening the pendulum P. 

Simpler forms of apparatus can be devised. For ex- 
ample, an ordinary metronome can be employed, the 
pointer passing behind an opening in a shield, the opening 
having graduations so that the position of the pointer 
behind the shield can be easily read. The sounder, which 
is always a part of the metronome, can be related to differ- 


Fio. 125 

From Wundt's " Grundziige der physiologischen Psychologie 
5th Ed., Vol. Ill, p. 82 

Fio. 126 


ent points on the scale by tipping the metronome into 
various positions so as to give the sound at various parts 

of the scale. The experi- 
ment consists in requiring 
the observer to determine, 
as before, at what point on 
the scale the sound is pro- 

A complication instru- 
ment designed by Prof. A. 
H. Pierce and described by 
him as follows, serves the 
purposes of this experiment 
very well: "In Fig. 126 
AAAA are card-holders at 
the ends of rigid arms, 90 

apart, attached to the main axis of rotation. BBBB are 
light steel rods, tipped with platinum wire, attached to a 
collar which rotates on the main axis behind the card-car- 
rying arms, and which can be set at any point by a screw. 
The platinum tips pass 
through a meniscus of mer- 
cury, the latter being con- 
tained in a well in the wooden 
block C. Thus a sounder, or 
telephone, is electrically actu- 
ated. In practice, the B's 
are somewhere behind the 
;l's. The cards to be placed 
in the holders are graduated 
as desired." 

"Fig. 127 shows the front 

screen in place. This is all that is seen by the observer. 
A vertical thread is placed in the window at the exact 

FIG. 127 


point where the sound comes. The subject notes the 
mark on the card which seems to be immediately behind 
this thread at the instant of hearing the sound. The 
actual objective simultaneity is determined empirically." 
Certain simple forms of exposure apparatus are de- 
scribed on page 197. These can be converted into de- 
vices which first expose and then cover up what they have 
exposed, by placing the figures or light so that they will 
be uncovered, not at the end of the movement, but in the 

Fio. 128 

middle of the movement. Other forms which may be 
mentioned consist of long pendulums which carry screens 
with openings of any desired length. Such pendulum 
tachistoscopes are open to the objection that they distract 
the observer by their movement. A form of tachistoscope 
which is free from the objection of a moving surface has 
been described by Professor Dodge. It consists in a box 
represented in Fig. 128. This box has two openings at 
W l and W 2 through which light from a convenient source 


may be allowed to pass. The light from W 2 strikes a 
mirror M 2 and is reflected to a surface O 2 , which is one 
field upon which letters or numbers or other desired ob- 
jects may be exposed. The light from O 2 passes along 
the line 2 F to the opening in front of which the observer's 
eye has been placed. In the course of the line FO 2 is 
placed a piece of dark glass TS. This dark glass is not 
as transparent as ordinary glass and reduces somewhat 
the light from O 2 . Nevertheless, it is sufficiently trans- 
parent to give an adequate light from the field O 2 and its 
chief purpose is served, not in connection with the light 
that comes through the opening W 2 , but in connection 
with the light which comes through the opening W 1 . If 
now the light is allowed to enter the box through the open- 
ing W 1 rather than through the opening W 2 , it will be re- 
flected from the mirror M l to the field O 1 , and from O 1 
the light passes to the dark glass TS. From the surface 
of the dark glass, the light is reflected into the eye of the 
observer. A certain amount of light also passes through 
this dark glass to the back surface from which it is in turn 
reflected into the eye of the observer, but it loses so much 
intensity in its double passage through the dark glass 
that the image from the back surface is practically lost and 
does not attract the attention of the observer. The field O 1 
is by this means directly superimposed upon the field O 2 . 
By an adjustment of the openings W 1 and W 2 the inten- 
sity of the two fields may be made equal to each other. 
A suitable device may be provided for changing the illu- 
mination from W 1 to W 2 , or in the converse direction. 
For example, a pendulum may be allowed to swing in 
such a way that an upper shield will cover the opening 
W 1 for a time, and a lower shield which is allowed to pass 
by W 2 will allow the light to enter through W 2 for any de- 
sired length of time. As the pendulum swings further, 


the lower shield may be made to cover the opening W 2 \ 
while the upper shield is opened so as to allow the light 
to enter through W l . While the light is coming only 
through the opening W l , the field O 1 will be in clear view 
and the field O 2 will be invisible. As soon as the entrance 
of the light is changed by the movement of the pendulum, 
the fields are reversed and O 2 comes into view to the ex- 
clusion of O 1 . The appearance of the fields in this case 
is without visible movement at any point, and the chief 
objection to the various forms of fall-tachistoscopes is 
entirely obviated. 

Experiments which require series of sounds given with 
regularity at short intervals can be set up with a metro- 
nome. They can, however, be properly carried out only 
with the aid of complex apparatus. Fig. 72 (page 145) 
showed an attachment for the Ludwig Kymograph for 
the purpose of utilizing that apparatus to make and 
break electric circuits which in turn operate sound ham- 
mers or spark devices. The wheel shown at the left of the 
figure is connected with the shaft of the clock-work. Con- 
tacts may be distributed along the graduated scale at 
known intervals, and from these contacts electric currents 
may be carried at desired intervals to properly arranged 
stimulating devices. 

A simple form of apparatus which serves many purposes 
in giving successive stimulations is a heavy pendulum 
swinging past a long graduated scale such as that shown 
in Fig. 122 (page 226). The scale should be graduated 
into time units. 

An elaborate apparatus for working with time inter- 
vals was designed by Schumann. This apparatus is 
driven by means of the Helmholtz rotation apparatus de- 
scribed on page 146. It consists of a graduated disk 
upon which are fastened a number of contacts, one or 


more of which can be easily moved so as to regulate in 
the course of the experiment the interval between con- 
tacts. A central arm connected with the rotation appa- 
ratus moves over these contacts at a regular rate, and the 
current thus made and broken is conducted to a sound 
hammer or other suitable device, where it is transformed 
into a stimulus for one or the other of the senses. 

The apparatus for recording the efforts of the subject 
to reproduce such a series of stimuli is the same as that 
required for the record in connection with the equivocal 
figures in the last exercise. If it is desired that the repro- 
duction of the series of stimulations shall be complete, not 
only a key and marker should be used, but provision should 
be made for the reactor to hear a series of sounds which 
correspond to his own movements. This series of sounds 
can be secured by connecting a sound hammer with the 
circuit between the reaction key and the marker. 


In general it will be found that the sound produced 
in a complication series is always referred to a wrong 
point in the visual series. Indeed, the experiment was 
suggested by the fact that the older astronomical observa- 
tions in which the observer was required to compare a 
visual and auditory series, always showed more or less 
error, so that sound and light were never reported in their 
true objective relation. No general statement can be made 
with regard to the type of error which will be committed 
in the experiments, as the direction and extent of the error 
depend very largely upon individual tendencies. 

The other experiments show the increase in scope of 
consciousness through grouping of experiences, and the 
impossibility of recognizing correctly a large series of 
impressions either when given simultaneously or succes- 


sively. In broad general terms it will be found that the 
range of accurate recognition is limited to from six to 
ten impressions, if these are given in such a way as to 
render rhythmical grouping impossible. 


The simplest experiment in time perception which can 
be performed is to give the observer a succession of three 
stimulations, the first two of which shall be separated by 
a standard interval, the second and the third by a variable 
interval. If this variable interval is successively increased 
and decreased, a determination can be made of the limits 
of accurate recognition of intervals. It will also be found 
by a similar experiment that the accentuation of any one 
of the stimuli in the series modifies the estimation of the 
interval preceding and succeeding the accentuated stimulus. 

If it is desired to use a continuous sound it will be 
found possible to produce such a sound by shunting an 
electric tuning-fork into a telephone circuit. The result 
will not be a musical tone in the telephone, but rather a 
burr which results from a rapid succession of makes and 
breaks in the electromagnet in the telephone. It is ex- 
tremely difficult to control a continuous sound of any 
ordinary type. The sound of a tuning-fork is not loud 
enough to act upon a telephone in a remote room and be 
transmitted as a musical note to the telephone near the 
observer. Any other form of transmission than the 
electric suffers from the impossibility of instantly shunting 
off the sound. There must be some continuous medium 
between the source of the sound and the observer in such 
a case, and this continuous medium almost invariably 
conducts enough of the vibration so that the observer is 
more or less distracted before and after the strong stimu- 
lus which is the subject of the investigation. 



The apparatus for this experiment can be made very 
simply. A strip of black cardboard may be passed 
through two openings lying in the same horizontal line in 
a white cardboard. Instead of a strip of black card- 
board some flexible material, such as a coarse thread, 
may very easily be employed; or a narrow, firmly woven 
braid will serve the purpose very well. Extending ver- 
tically across the portion of the black strip or thread 
which is visible to the observer, should be the short 
cross-line described in the text. The observer can now, 
by reaching behind the white cardboard, draw the black 
strip with its vertical cross-line either to the right or to 
the left, and in this way adjust the cross-line at any de- 
sired distance from the points where the line is cut off by 
the cardboard. 

A second method of setting up this experiment is to use 
the apparatus described under Exercise I for the measure- 
ment of illusions. A horizontal line is drawn on the fixed 
card and a short black line is exposed from behind a white 
screen, the cross-line and the screen being attached to the 
movable board in the apparatus. The short cross-line 
is now adjusted to the desired position along the long 
horizontal line. The cross-line may be drawn on the 
back of a glass, in which case there will be no lines in the 
field of vision except those which are to be compared. 

For the second part of this experiment, the simplest 
method consists in laying before the observer a strip of 



black paper upon a white background. A second strip of 
similar paper is provided, and the observer is required by 
means of a white screen to cover up such a portion of this 
second strip as he may desire, and to place the visible por- 
tion of the second strip across the first piece that was pro- 
vided. Here again lines on the back of glass may be used. 
For the third part of the experiment a series of figures 
cut from black cardboard or from colored paper may be 
used to counterbalance the vertical line which is drawn 
upon the paper at the outset of the experiment. These 
can be readily adjusted by the observer until he is satisfied. 
Measurements' should be made in this case to the edge of 
the figure which is nearest to the central point in the field 
of vision. A very good series of figures for this purpose 
can be made up by drawing rectangles; first, five which 
are 20 mm. wide and successively 1, 2, 4, 8 cm. long; and, 
second, five which are each 4 cm. long and successively 
1, 2, 3, 4 cm. in width. Circles of various sizes are also 
easily produced. 


The formula which is most common for unequal di- 
visions is that which has commonly been called the golden 
section and is as follows: The long part is to the sum of 
the two parts as the short part is to the long part. This 
principle holds also for the division of crosses and for the 
relation between their two legs. 

The results of symmetrical arrangement of figures is 
a matter which depends on the size of the figures. Gen- 
erally speaking a large figure near the center balances a 
small figure far away. 


The supplementary experiments require no comment, 
except possibly the last, which refers to the method of 


selection. This method consists in presenting to an ob- 
server two colors or forms, and asking him to select the 
one which is more agreeable. Other combinations are 
now presented, including the selected color or form, until, 
finally, by a succession of choices the observer has arrived 
at a final selection of that one which seems to him to be 
most agreeable. This method was suggested by Fechner 
in "Vorschule der Aesthetik." 


For the experiments with these higher mental processes 
the physical conditions are not elaborate. The appa- 
ratus may be of the simplest sort. In the two experiments 
described, the apparatus consisted of a projection lantern 
which would throw images upon the wall in front of the 
observer, and of a simple light which could be turned on 
the wall in very faint intensities. 






Drawing material: 

Drawing board. 

Drawing paper. 




Ruling pens. 


India ink. 
Carpenter's tools. 

Electric current. 

Table clamps. 



Meter rods, tapes and rulers. 

Projection lantern with slides. 
Photographic equipment for 

making slides. 
Supply of millimeter paper. 

Metal and wood-working ma- 
chine shop. 

Lamp batteries. 

Plugs and sockets for connec- 



Sets of cards, five pairs in each 

Paper on which to mark lengths 
of lines as set. 

Holder for cards during adjust- 

Millimeter measure. 


Apparatus for holding cards 
with recording attachment 
and roll of paper on which 
to make records. 

Thumb-tacks to fasten cards to 




Head-rest (table clamps, Head-rest (independent). 

S-clamps, rods.) 
Shield for eye. 
Colored papers. 

Perimeter or campimeter of Perimeter of more elaborate 
simple construction. type with light box for ex- 

posure of colors. Electric 



Colored papers. 
Colored disks. 

Simple color disk rotator. Electric motor with arbor for 

Protractor. holding disks. (Current con- 

nections, wires.) 



Head -rest (table clamps, Head-rest (independent). 
S-clamps, rods). 



Meter rod. 

Table clamps, S-clamps and 
rods to hold measuring rod. 

Paper ruled in large squares. 

After-image card. Box with lamp and Aubert's di- 

aphragm. Dark room. Elec- 
tric connections. If these are 
not at hand the third part of 
the exercise must be omitted. 



Model of truncated pyramid. 
Drawing material: 
Paper, ruler, etc. 



Mirror stereoscope of simple Adjustable mirror stereoscope, 

(Lens stereoscope may be 



Head-rest (clamps on chair). Head-rest (independent). 


Two sounders. Two telephones, mercury key, 

electric connections, battery, 


Auditory cage, simple construe- Auditory cage, adjustable, 



Quincke's tubes. 

Mouthpiece for blowing same 

in combinations. 
Two chromatic pitch-pipes with 

scale. Set of tuning-forks. 

Resonators (one adjustable). 
Conducting tubes leading from 



Ink pad. 

Metallic points in cork. Holder for various kinds of 

Wooden rods, sealing-wax, bris- points, 

ties of various sizes. Metallic points. 

Bristles and grips to fit holder. 



Wooden stylus. 
Aniline ink. 

Compass with hard rubber or Aesthesiometer with scales, 
bone points. 



Millimeter measure. 
Pasteboards cut into various 


Hard rubber strips with edges of 
various lengths. 




Rotating table for presentation 

Pith-ball sounder (simple con- 
Photometer (shadow). 

of weights to observer. 
Sounder with accurate holder 
for balls. 




Rubber dam, thread, beeswax, 
rubber tubing, straws, mate- 
rial for points. 

Kymograph. (If no motor is 
provided in equipment, clock- 
work kymograph). 

Kymograph paper, paste. 
Flame for smoking paper. 
Shellac tray or table. 
Electric fork with connections. 



Thistle tube. 

Clamps for above. 

Rubber tubing. 

Recording tambour. 

Kymograph and accessories. 

Time marker. 

Sugar solution, quinine solution 

Air valve. 

Two drums. 

Pulleys for connection with 

Motor and connections. 

Marker for use with fork. 
Interrupter for longer intervals. 
Pendulum with contacts for 

long intervals. 
Coarse marker recording with 

ink or pencil. 
Jacquet chronometer. 






Tambours (receiving and re- 

Kymograph and accessories. 
Planchette with recording pen. 



Spring key with electric connec- 


Kymograph and accessories. 

Time marker. 

Hand-rest to bring the reactor's Apparatus for holding paper 
hand over a moving paper on over traveling paper. Type- 

which he may draw the writer ribbon or carbon paper 

straight lines required. underneath. 



Two keys (experimenter's and 


Electric tuning-fork and connec- 

Ewald chronoscope and wires 
for connections, or Hipp 
Electric markers. 

Tambours (one of these may be Lever key for reaction forms, 
substituted for the reactor's Sounder to give signal, 
key). Shutter to expose colors. 

Kymograph and accessories. 



Three keys (experimenter's, two 

for reactor). 

Electric fork and connections. 
Markers. Chronoscope and connections. 



Tambours. (One may be sub- Lever key for recording form of 

stituted for the reactor's key.) reaction. 
Kymographs and accessories. 


Color exposure apparatus con- Exposure shutter. 

nected with experimenter's Articulation key. 




Attachment for kymograph to 
take record of writing. Apparatus, as described by Mr. 

Recorder to attach to hand. Freeman, for writing record. 

Figure for moving point on ky- 

Carbon paper to record hand Cylinder for recording band 
movements. movements. 



Series of patterns. 

Paper for drawing. 

Carbon paper. 

Soft paper. 

Metal stylus. 

Millimeter ruler. 

Protractor to measure angles. 



Spring key with connections. 


Time marker connected with 

fork or interrupter. 
Kymograph and accessories. 

Watch. Stop watch. 

Cards marked with numerals or 




Spring key with connections. 



Kymograph and accessories. Porter's ergograph. 




Cards with nonsense syllables. 

Kymograph and shield for ex- 
posure of syllables. 

Ratchet for intermittent move- 
ment of kymograph. 

Wirth's special apparatus. 



Equivocal figures. 

Key, marker, kymograph and 

accessories, time-marker. 
Stereoscope, colored papers. 
(Masson disk, color mixer). 



Metronome and shield. Complication pendulum. 

Exposure apparatus and letters Dodge's exposure apparatus. 

and figures for same. 
Metronome for series of sounds. Pendulum with contacts and 

sounder, or 

Spring key, marker, kymograph Attachment for Ludwig Kymo- 
and accessories. graph with contacts and ham- 





Adjustable lines. Illusion apparatus, adjustable 

Millimeter ruler. cross-line on glass. 

Strips of black paper. 

Black card figures for symmetry. 




2. Analyze white light into components with the aid of a 

prism or spectroscope. Or mix colors and produce gray . 47 
6. Memory experiments as illustrating method 222 

11. Plotting of blind-spot 42 

15. Chart or lantern slide of amoeba or other cells. 

17. Other charts from Jennings. 

23. Models, or charts, or lantern slides of nervous system of 
various invertebrate forms. 

27. Preparation or model of frog's nervous system. To make 
the preparation, fine, sharp-pointed scissors, a scalpel, 
and forceps are required. The frog should be chloro- 
formed and dissected. 

30. Models or preparations of vertebrate brains higher than 

36. Model of human head showing brain in situ. 

36. Model of brain showing sections. (Azoux brain model is 
the best.) 

36. Charts or slides showing various types of cells and various 
parts of the nervous system in fine anatomy. 

36. Charts by Strumpell and Jakob, published by J. F. Leh- 
mann, Munich, Germany. 

75. Spectrum (if not shown in connection with page 2). 

75. Gray and colored papers 36 

77. Chart of color circle. 

81. Model of eye, showing parts. 

84. Kiihne's eye, or other means of demonstrating properties 
of lenses (camera obscura). 

84. Show properties of monocular image 64 

84. Thread model showing paths of rays of light entering the 

84. Model showing muscles of eye. 




89. Charts of retina when dark-adapted and light-adapted. 

90. Color-blindness, normal and abnormal. 

93. Color mixing 47 

96. Pigment mixing 61 

96. After-images 66 

97. Color contrasts 58 

103. Appunn's reed to demonstrate origin of sound, also 

lowest tones 117 

103. Galton's whistle, or metal bars of highest tones 117 

104. Records of sounds 106, 214 

105. Model of ear, showing parts. 

109. Experiments on bodily localization 126 

113. Beats and combination tones with adjustable forks and 

Quincke's tubes 106 

117. Model of median section of head. 
117. Charts of sensory surfaces. 

120. Taste experiments 120 

123. Analysis of tactual qualities 118 

125. Model of skin, showing parts. 

128. Weber's Law. (The photometer demonstration is best) 128 
131. Functional effects of sensations. (Demonstration better 

138. Tactual space 121 

144. Auditory space. (Simple demonstration with snapper 

sounders) 98 

148. Miiller-Lyer and other illusions. (Methods of meas- 
uring) 16 

154. Retinal image and objects 64 

157. Binocular depth 78 

160. Stereograms and lens stereoscopes 93 

160. Monocular flatness 68 

161. Retinal rivalry 86, 229 

175. Time experiments. (Simple series of taps may be made 

by the instructor to test the range of recognition) .... 236 

184. Dynamometer 214, 221 

185. Sphygmograph, or plethysmograph 160 

186. Planchette 169 

189. Attention 220, 229, 231 

198. Forms of simple reactions 182 

205. Aesthetics. . . 239 



206. Illusion of weights 172 

209. Tonal fusions 113 

211. Reproduction of tones 214 

219. Diffusion and organization of habits 169, 217 

221. Writing and other coordinations 176, 206 

231. Memory 224 

236. Memory in broad sense 213, 217 

317. Throughout this chapter see Exercises XII XX. 

321. Reflex winking 219 

322. Perceptual activity 169 

323. Form of action 192 

334. Reaction experiments 182 

358. Aesthetical appreciation 239 



Accommodation, visual. . .73, 84 

^Esthesiometer 122 

^Esthetic appreciation 239 

After-image, tactual 123 

After-image, visual 66 

Air pressure 107 

Air valve 142 

Algometer 120 

Angell 199 

Apparatus, lists of 243 

Appunn Ill 

Arm rest 162 

Articulation 214 

Articulation, reactions 203 

Association, free 204 

Association, reactions 203 

Association, restricted 204 

Attention 229 

Aubert's diaphragm 66 

Audiometer, telephone 129 

Auditory cage 101 

Auditory localization . . . 104, 105 

Balance, aesthetic 240 

Batteries, electric 4 

Battery, lamp 6 

Beats 116 

Belt kymograph 147 

Bergstrome 223, 226 

Binaural perception 98 

Binocular vision 78 

Blind-spot 42 

Bodily localization 126 

Boekel & Co 109 

Brake for electric motor ... 53 

Breese 95 

Brightness contrast 58 

Brightness equation 37 


Cage, auditory 101 

Cameron 214 

Campimeter 35 

Capillary pen 157 

Card-distribution test 217 

Charts, demonstration 2 

Charts, visual tests 76 

Chronoscope, Ewald 183 

Chronoscope, Hipp 187 

Chronometer, Jaquet 153 

Circle illusion 25 

Circular measurements .... 15 

Circulatory activities 160 

Clamps 9 

Color absorption 61 

Color blindness 41 

Color-blind test 56 

Color contrast 58 

Color equations 56, 61 

Color mixer 51,52,54 

Color mixing 47 

Color top 49 

Colored papers 36 

Commutator 190 

Compass 121 

Complex mental processes.. 242 
Complication experiments. . 231 

Confusion experiment 123 

Connection for Ewald Chro- 
noscope 184 

Connection for Hipp Chro- 
noscope 188 

Contact apparatus 199 

Contrast, brightness 58 

Contrast, color 58 

Contrast disks 59 

Control hammer 190 

Convergence, visual 88, 96 




Coordinate aper 


nate paper ......... 12 

Countershaft ............. 146 

Cutaneous qualities ....... 118 

Cylinders for lowest tones. . 116 

Dark adaptation of eye, 45, 46, 60 
Demonstrations ......... 1, 250 

Depth, monocular ........ 71 

Depth, perception of ..... 68, 69 

Diaphragm, Aubert's ...... 66 

Diedrich ................. 133 

Difference tones .......... 116 

Direction, tactual percep- 
tion of ................ 124 

Discrimination reactions . . . 203 
Disks, color .............. 49 

Distance, auditory ........ 98 

Distance, perception of .... 72 

Distraction .............. 220 

Division, aesthetical ....... 239 

Dodge .................. 234 

Double images ........... 78 

Drawing ................. 3 

Drawing reaction ...... 175, 213 

Dynamometer ............ 172 

Edelmann ............... 117 

Edison-Lalande batteries . . 4 

Electric motor ............ 52 

Electric plug ............. 7 

Electrical connections ..... 3 

Equation for retinal image. 65 

Equivocal figures ......... 229 

Exposure apparatus. ... 196, 234 

Exposure shutter ......... 198 

Eye shield ............... 31 

Fall apparatus, sound. . 129, 132 
Fall tachistoscope ......... 197 

Fatigue .................. 221 

Fechner ................. 241 

Filled space illusion ....... 26 

Fluctuation of attention. . . . 229 

Form of reaction .......... 192 

Freeman ......... 176, 180,206 


Friction clutch 177 

Fusion, tonal 115 

Galton's whistle 117 

Gelatine 40 

Golden section 240 

Graphic representations ... 12 

Gray paper 44 

Habit 217 

Hammer, electric 189 

Hand movement recorder. . 211 

Hathaway 2 

Head rest 31 

Helmholtz' rotation appa- 
ratus 146, 236 

Henderson 166 

Henri 124 

Hering 36 

Illusion apparatus 18, 28 

Illusion of weights 172 

Illusion, visual 16 

Initial threshold 11, 128 

Interrupter, electric 99 

Interrupter, Kronecker. . . . 156 

Introduction 1 

Inverted shadows 70 

Involuntary movements. 160, 169 

Jaquet chronometer 153 

Jastrow 171 

Keyboard, organ 109 

Kinetoscope 88 

Kohl 87, 110 

Kronecker, interrupter 156 

Kymograph 143 

Kymograph record 158 

Lamp battery 4 

Lampblack 150 

Lantern, for demonstrations, 2 

Laryngograph 171 

Light adaptation of eye. ... 45 




Light-box, for perimeter ... 39 

Light contrasts 58 

Lines, tactual perception of, 124 

Localization, auditory 104 

Localization, bodily 126 

Localization, tactual 121 

Loomis 174 

Lowest tones 116, 117 

Ludwig kymograph .... 145, 236 

Marbe.. 54 

Marker, electric 155 

Marking-out test 204 

Masson disk 230 

Memory 234 

Mercury contacts 157 

Metronome 220, 231 

Meyrowitz 38, 39 

Milton-Bradley 37, 51 

Minimum visibile 75 

Mirrors, single image 84 

Mixer, color 51, 52, 54 

Models, demonstration .... 2 

Monochord 228 

Monocular adjustments. ... 175 

Monocular depth 73 

Monocular vision 64, 68, 86 

Mosso's ergograph 221 

Motor, electric 52 

Movement, tactual percep- 
tion of 125 

Miiller-Lyer illusion . . 12, 16, 22 
Miinsterberg 51 

Nonsense syllables 224 

Odors 133 

Olfactometer 133 

Organ pipes 109 

Pain 120 

Paper-cutter for disks 50 

Pen, universal 170 

Pencil marker 157 

Pendulum, complication . . . 231 


Pendulum, contact 226, 236 

Pendulum, sound 133 

Perimeter 33 

Peripheral vision 34, 41, 45 

Peripheral vision of form . . 44 

Personal equation 237 

Personal factor in experi- 
ments 11 

Photometer 131 

Photometry 45 

Pierce 105,233 

Pigment mixtures 61 

Pillsbury 124 

Pifch-pipe, chromatic 106 

Pith-ball audiometer 129 

Planchette 169 

Plethysmograph 164 

Plug contact 7 

Pneumograph 167 

Poggendorff illusion 24 

Porter 137 

Porter's ergograph 222 

Porter's kymograph 144 

Practice curve 13, 14 

Practice results 217 

Practice tests 217 

Preface i 

Pressure recorder 178 

Pressure spots 119 

Pressure threshold 135 

Procedure with knowledge. 10 
Procedure without knowl- 
edge 10 

Projection, visual 64 

Protractor 50 

Pseudoptics 51 

Pseudoscope 91 

Psychology, General Intro- 
duction 1, 14,80, 171 

Pupilar reaction 168 

Quincke's tubes , 106 

Reactions, complex 201 

Reactions, simple 182 



Reaction key 193 

Reaction, writing 206, 208 

Recording points 141 

Recording tambour 138 

Reflex winking 219 

Relativity, temperature .... 119 

Resonators 113 

Retinal image 65 

Retinal rivalry 86, 95, 230 

Rotating table 127 

Rothe 36,52 

Rousselot 171, 172 

Sanford 191 

Scripture 4, 191 

Scripture's touch key 199 

Seashore 103 

Section, golden 240 

Semicircular canals 126 

Sensation, intensities of. ... 128 

Shellac bottle 151 

Shellac tray 150 

Schumann 236 

Size, visual perception of . .66, 71 

Smith, C. H iii 

Snapper sounder 98 

Socket, electric 8 

Solar prints 2 

Solidity 80,87 

Sonometer Ill 

Sound hammer 189 

Sound pendulum 133 

Sound threshold 128 

Spectroscope 37 

Spectrum 47 

Sphygmograph 160 

Sprague 2 

Spring key for tapping 175 

Stamp for skin 118 

Standard 159 

Standard Dictionary 38 

Stereograms 81, 94 

Stereoscope, lens 92 

Stereoscope, mirror 82 

Stereoscope, Wheatstone ... 9 

Stern Ill 

Stern's tone-variator 134 

Stocking Co 36, 37, 44, 99, 


Stratton 91 

Stroboscope 87 

Stumpf 116 

Symmetry of figures 239 

Table-clamp 9 

Tactual localization 121 

Tactual threshold 125 

Tambour 137 

Tambour pen 212 

Tapping 175, 181,220 

Target experiment 215 

Taste 120, 134 

Teeth-rest 33 

Teeuwen iii 

Telestereoscope 92 

Temperature spots 118 

Thistle-tube sphygmograph, 160 

Threshold, initial 11, 128 

Threshold, pressure 135 

Threshold, sound 128 

Threshold, tactual 125 

Time, empty 238 

Time experiments 236 

Time, filled 238 

Titchener 108 

Tonal fusion 115 

Tonal intervals 115, 116 

Tonal perception 106 

Tone conductors 112 

Tone-messer Ill 

Tone-variator Ill 

Tongue bulbs 172 

Tuning-fork 112 

Tuning-fork, electric 153 

Tuning-fork holder 155 

Typewriter test 219 

Underwood & Underwood . 94 



Valve, air-pressure 109 

Vaso-motor reactions 164 

Verdin 140, 171 

Vision, acuity of 76 

Vision, binocular 78 

Vision, chart tests 76 

Vision, monocular 64 

Visual distance 66 

Voice key 201 

von Frey 122 

Watch test for ear 130 

Weber's Law 128 

Weight illusion 173 

Weights, psycho-physical 
tests. . . .128 


Wheatstone stereoscope 9 

Whipple 108,205 

Whisper test for ear 130 

Winking reflex 219 

Wirth 227 

Witmer 58 

Writing reactions 206 

Writing recorder 177 

Wundt 112, 113, 132, 185, 

186, 201, 231, 232 

Zimmermann 135, 144, 145, 

146, 156, 165, 189, 190, 225, 

Zollner illusion . . 27 




* 1 7 ,94, 

CWjl ;,T 03Q 


i OCT 92 1Q47 


| w - 

MN(4 ^ 

JMIM 2 4 1954 

OFCtS ' . 


JUN 2 8 1960 


IAII 1 R 1962 


c r 7 'fior i 

LD 21-100jn-8,'34