• Users Online: 1987
  • Home
  • Print this page
  • Email this page

   Table of Contents      
Year : 1962  |  Volume : 10  |  Issue : 2  |  Page : 27-35

Modern photopic perimetry

Paris, France

Date of Web Publication29-Mar-2008

Correspondence Address:
A Dubois Poulsen
Login to access the Email id

Source of Support: None, Conflict of Interest: None

Rights and PermissionsRights and Permissions

How to cite this article:
Poulsen A D. Modern photopic perimetry. Indian J Ophthalmol 1962;10:27-35

How to cite this URL:
Poulsen A D. Modern photopic perimetry. Indian J Ophthalmol [serial online] 1962 [cited 2021 May 14];10:27-35. Available from: https://www.ijo.in/text.asp?1962/10/2/27/39668

It is unnecessary to state how useful is perimetry in ophthalmology and neurological clinical work.

It represents one of the few, if not the only method, able to detect the state of intracranial optical pathways, and it allows precious observations in cases of affections more specifically ocular. Apart from the new technics,-perimetry in darkness, perimetry at mesopical levels, perimetry after dazzling, flicker perimetry, one of the most fruitful ideas seems to have been to replace the former perimetry with single variable stimulus by two variable ones, which are intensity and test surface.

Let us state what older perimetry consisted of. Methods were based on the notion of test visibility in the visual field and sensibility gradients from center to retinal periphery.

Visibility of an object, whatever it is, depends on numerous factors of which the most important are:

(i) its surface area (ii) its lumin­ous intensity and (iii) its con­trast with the background upon which it is presented, that is the luminous level of this background.

Moreover it depends on its excentricity in the visual field, in other words on the retinal region upon which its image is produced.

To give these different factors their accurate value, one has endeavoured, as in every physical experiment, to vary only one para­meter at a time, the other remain­ing constant.

The experiment has been then re­peated as many times as necessary, giving different but fixed values to secondary parameters. Families of curves have thus been deduced which are most valuable in visual physiology.

To vary only one parameter at a time evidently is the simplest and easiest idea. The most directly accessible variable is the test sur­face. Therefore, in clinical work, RONNE'S wellknown technic of the black or grey background with con­stant lighting and white tests with variable and known dimensions, has been adopted. TRAQUAIR, whose book includes every conclu­sion possible, brought that technic to its utmost perfection.

The larger the test, the more peripheral the site in which it can be observed in the visual field. Sensibility can be defined as re­presented by the inverse of the test size. So, carrying excentricities in abscissa the fixation point being quoted O, a characteristic curve of retinal sensibility from center to periphery can be obtained. On 3 and not 2 dimensions TRAQUAIR'S famous "island, washed in a sea of blindness," is produced, characteris­ed by the central peak of foveal vision, close to which is found the crater of Mariotte's blind spot. The shore slopes are more or less inclin­ed according to the meridian of the visual field examined.

This, introduced in perimetry the isopter notion that is to say level curves joining heights of equal sensibility on the island, thus trac­ing closed curves in the visual field.

In pathology the pattern of curves and their gradients must be studied. TRAQUAIR'S conclusions are well known about the early damage of central or peripheral isopters according to the lesion observed as well as his considerations on soft slopes of evolutive or unstable lesions in opposition to the steep slopes in the case of fixed or stable one:3.

This is a very elaborate technic. It has allowed accurate diagnosis. Recently HARMS substituted the surface variation of the test by the luminous intensity one. On every point of the visual field the lumin­ous intensity is increased, all other variables being kept constant until visibility is reached. In fact it is a research for local thresholds, but still proceeding from the one van­able parameter idea.

Perimetry is in fact meant to search for thresholds through quicker and coarser clinical pro­cedures than physiological methods. It really tends to find in the visual field the points corresponding to a given threshold defined by the qualities of the chosen test. There­fore, there is no objection to basing perimetric technic upon physiologi­cal principles of research for thres­holds. Now, these are depending on numerous parameters the variations of which are not independent but interlinked by laws.

To make use of these interaction laws is the object of new perimetry. To thrive on new signs clinical work calls on theory.

  Spatial Summation Top

Threshold variation factors are exactly similar to visibility factors above mentioned, i.e. test angular dimensions, luminous intensity, position or eccentricity, colour, time of exposure, contrast with back­ground, all of which cannot evident­ly be considered. Laws linking test surface or its luminous intensity to TIME have not yet been considered in clinical perimetry where the test has always been presented with limited fixed time. So far COLOUR is not a factor in the examination because of our lack of physiological knowledge on its peripheral percep­tion. It is usual practice to consider therefore white tests only and laws linking the luminous intensity of the test to its surface. Those are named laws of spatial summation . Let us define them briefly, state the physiological and quantic interpreta­tion given and establish the way in which they can be used in clinical work and then in reverse see how clinical work can, in turn, help to understand them. INTENSITY AND SURFACE being two of the main threshold variables, it is legitimate to qualify the new perimetry of two variable technics, to contrast it to the former one of one variation factor, all the others remaining constant during the experiment.

Relation between luminous in­tensity and test surface area must be established both for luminous absolute threshold and differential threshold.

ABSOLUTE THRESHOLD. By absolute threshold is meant the ability to recognise a test object varying in size or intensity on a uniformly lighted background.

Authors who applied themselves to threshold mensuration, do not always give the relation found between surface area and luminous intensity, but it is implicited in their results and can in any case be deduced from their given curves ex­pressing intensity in function with surface.

Others have expressed through a mathematical law the relation between the two quantities. Accord­ing to the former results there should be a constant quantity of light necessary to reach the sensibility threshold and expressed by the well­known formula:

I S = const.

Where I is intensity and S is sur­face area.

It is RICCO'S law, dating from 1887, taken over again by CHARPENTIER and rediscovered later by LOHLEIN in 1905 and HENNIUS in 1909. But is this law valuable for the whole retinal surface ? According to GEHLOER and SCHERRING (1919) RICCO'S law is accurate for peripheral retina but not for fovea. At the latter level constancy is assured by the product of intensity by surface square root.

In 1903, however, PIPER stated that under the I√S = const. formula the law could be applied to the whole retina hence the name of PIPER'S law given to this new for­mula.

PIERON, in 1929, reconsidered the question in a stricter manner with the head immobilised and a fixation point for the eyes very carefully studied.

The curves expressing relation between surface and luminance are exponential functions whatever the wave length of light and expressed by equations of the form:

I S k = const.

where k is called summation ex­ponent.

PIERON proposes I S 0.33 = const.

or I 3√S = const. ELSBERG and SPOTNITZ, American authors have taken it up again in 1937, much later than PIERON to find again the I S 0.33 = const. law, which ever since has been called after them. GRAHAM Brown and MOTE, and BAUMGARDT found again similar figures.

Nothing seems more perplexing than this research of relations between surface and intensity. There are three possible laws. All have the I S k = const. formula but the exponant k varies according to authors. Thus it is unit for Ricco, square root for Piper, cubic root for Pieron and it is even possible for the exponent to vary still more and reach intermediary values between 1 and 0.7, e.g. 0.9, 0.8 etc.

Variables interfere in various ways. Variations in test size may, for instance, condition the form of equation, With a small surface, the exponent may equal one, with a larger surface it may take smaller values than I, for instance 0.5 or even to abolish itself. Similarly, luminance may introduce modifica­tions and finally test excentricity is itself a prevailing factor. Diameter of the pupil, test colour and optical imperfection factors also play their part.

It is sufficient, to make use of these laws in clinical work, to settle the limit of their validity and pick out the simplest. After considering the problem at length, we prefer to rely upon a simple and constant re­lation between test luminance and surface size, for whichever region of the retina that is to be explored. One must adopt test-objects smaller than 1 ° in surface-size and not higher than 10-9 stilbs (between 10-9 and 10-7) in luminance. This relation holds better in the retina more peri­pheral to 30' from the fixation. These figures should be the limits for clinical exploration for absolute thresholds.

DIFFERENTIAL THRESHOLDS : By the differential sense is meant the ability to recognise two similar test ­objects as distinctly separate. The classical wav to search for differen­tial thresholds would be to recognise two contiguous luminous stimuli but because of the varying sensitivity in the different parts of the retina this method would introduce difficulties. It can be estimated in another way that is by searching for a differential fraction, that is to search for a threshold for recognising a small luminous object over a larger less luminous background. This may appear to be just the same as when testing with a luminous object on a uniformly lighted total visual field as while using a Goldmann's peri­meter, but we shall see the diffe­rence presently.

Author's curves concerning dif­ferential fraction expressing a cer­tain contrast between a test object and its lighted background may cor­respond to several test dimensions and several different lighting levels. Isopter method allows them to con­ceive spatial summation laws in an easier and practical way in clinical perimetry. In fact, one isopter cor­responding to a differential thres­hold may be registered on the same place of the field with other in­tensity and test-surface-area com­binations. A large contrast and a small dimension test-object may be equal to a weak contrast and to a larger dimension test-object. There is thus one more law for relations between surface and contrast. It is another spatial summation law.

There is thus a definite relation between size of object and contrast of luminance to produce the same visual response in the retina. It is possible to determine corresponding isopters for different combinations of size-dimensions and luminance­ contrast. The determination of these isopters constitutes the essence of PERIMETRY WITH TWO VARIABL­ES. When isopters determined by two different combinations of size and luminance which should normally correspond, do not cor­respond then there is a disturbance of the differential sense. We shall call it SUMMATION TROUBLE.

and RAND seem to have first tried to express with curves, relations between the visual field. dimensions and the two variables, but their system is heavy and diffi­cult to handle.

GOLDMANN'S system is coherent and practical in application. On the perimeter, named after him, a lamp insures the test luminous intensity and its variations are obtained by interposition of filters. He reaches a formula of form 1 S k =const. where K = 0.84, which, according to GOLDMANN is the best average. It is clearly shown that to obtain equal isopters with different sizes of tests, surfaces must increase in geometrical progression, while filter permeability must decrease in another progression.

These laws are only applicable within certain limits. The more peripheral the zone in the field and the larger the test, the better are summation phenomena.

It is necessary not to cover a surface over 1.1 mm 2 at 33 cm. that is to say once again 1◦, for the law to remain accurate for the whole field area.

We have, ourselves, gone through these experiments again with parti­cular attention to the formula ex­ponant. It is precisely the exponant which characterises the summation -1.0 according to RICCO'S law, 0.5 in PIPER'S law, 0.3 with PIERON'S. So we thought it necessary to deter­mine this exponant directly in a normal subject as well as in a pathological case. We have modifi­ed GOLDMANN'S apparatus, which, with its filters allows only a varia­tion of luminous intensity step by step. With GOLDBERG's wedge we have been able to obtain an un­interrupted variation. This apparatus allows the drawing of an isopter with a first test of given surface and intensity. Then, picking out another surface area one feels one's way about to find out what intensity to give to the test so that the isopter places itself in the same site of the visual field. I am not going to enter into details, however, very simple of logarithmic calculations.

With 5 individuals trained to physiological experimentation and observed on S isopters, one has been able to find an average ex­ponant of 0,952-almost 1 which corresponds to RICCO's law.

Sixteen untrained individuals have however allowed to find an exponant equal to 0.818.

The average between the two figures is 0.85, very close to 0.84, GOLDMANN's exponant. This 0.85 is quite a worth while clinical value in practice, but it is better to keep nearer stricter experiences and adopt 0.90.

This exponant varies in an anar­chical way through the visual field according to the excentricity of the isopter observed. But this variation is in keeping with the margin of errors and one may consider the exponant as constant, in the whole field area for the sizes of the tests examined. The central area upto 5 ° however has not been explored.

It is out of the question to use intricate calculations in clinical work.

The spatial summation relation is rendered quite easy with GOLD­MANN's apparatus. There is one number for each test surface and each intensity. It is enough to pick out the summated intensity--surface stimulus in such a way as to have the sum of their numbers constant in order to obtain the whole series of tests belonging to the same isopter. An extremely practical Pythagoras' table indicates all sur­face and intensity combinations. They are found on the diagonals of the table.

When curves thus defined by different but corresponding tests do not coincide, there should be some summation trouble . One must how­ever keep in mind that GOLD­MANN's chosen exponant is an average, constant only for minor tests and that summation is only really applicable at field periphery.

Normal isopters do not always coincide completely and it is better to consider only important variations over 5° .

It must be immediately em­phasised that differential sensibility cannot characterise alone the peri­metric properties of a retinal zone. It is necessary to determine the summation capacity characteristic of the observed area. HARM's em­bossed model or TRAQUAIR's schemes expressing sensibility variations in the name of only one variable are thus not sufficient enough. It is here that the applica­tion of two variable perimairy in clinical work gives a striking proof of this. Many perimetrical lesions escape in fact, to a certain test combination which can be detected by other combinations which rigorously equivalent in a normal sub­ject, become non-corresponding in a pathological case.

But we must not anticipate. We must briefly say what are, in ner­vous physiology, laws of corres­pondence between intensity and test surface.

  Physiological Spatial Summation in Retina Top

Phenomena studied through two variable perimetry are in reality the ones grouped under the name of spatial summation. There are differ­ent types of physiological summa­tions.

A lesser stimulus than the threshold does not elicit a response. However, if repeated at a suitable interval, it elicits an answer from the examined system. It is temporal summation in which influxes tend to add themselves in time. If several axon reflexes tend to converge on one single cell, in sufficient abundance it can be activated. This is another variety of spatial summation in which in­fluxes come from several points of space. In retinal physiology, when the surface of an infraliminary, luminous intensity test is enlarged numerous sensorial cells are excited, the influxes of which tend to con­verge on one or several ganglion cells. This creates an addition of in­fluxes on the ganglion cells and a spatial summation is realised. Heter­ogeneity of the retina in its different parts strongly conditions this pro­blem. At foveal level., each cone is joined to a bipolar cell which con­verges on a ganglion cell, summa­tion is thus perfect there. This type of summation is better seen when the stimulus is peripheral. Actually, a more complete interpre­tation of the facts can be given, based on statistical laws and cor­puscular theories of light. It is known that radiations have un­dulatory characteristics, their ab­sorption by matter takes place under quanta form. One quantum constitutes the smallest parcel of radiant energy capable of isolated existence. It is carried by photon. A weak radiation carries few photons. It is ruled by hazard laws and its statistical character is some­times richer, sometimes poorer. The well-known threshold fluctuation is then explained by fluctuation of weak intensity flux and it is not necessary to bring nervous system in this theory.

Having gathered the idea of unit of foveal conduction as opposed to group conduction in the retinal peri­phery one can explain the varying behaviour of those two retinal zones under weak lighting. Foveal conduc­tion units which are very small can pick very few photons strong enough to strike a sensorial cell and produce stimulus in a ganglion cell, while peripheric groups easily pick up larger num­ber of the weaker photons which are summated to produce a stimu­lus strong enough to cause stimula­tion of the common ganglion cell in which they terminate. Hence we find central scotoma in the dark and peripheric sensibility maintain­ed under cones zoons of weak illumi­nation (PIRENNE).

To explain summation phenomena one must refer to retinal histology notions where convergent disposi­tions of cells of one with another play their part. Therefore, the inference that summation in the problems of vision has a purely retinal substratum is but a short step, too easily taken by physio­logists.

  Geniculate Bodies and Occipital Cortex Top

Two additional organs may be present with retina centers for nervous reductions and conver­gences. They are the corpus geni­culatum and occipital cortex.

It is now necessary to make re­search in clinical work to reach a more perfect understanding of summation phenomena in the visual domain. We have given ample time to back high levels of theory to reach the lower regions of practice. Let us however take notice that from our point of view clinical work is not an intruder but an absolute scientific necessity.

To get a rapid idea of the value of the spatial summation trouble research, we have gone over visual fields measured in the Quinze-­Vingts hospital with Goldmann's perimeter (about 3000 fields).

We have made two separate lists of cases : the former concerning those with summative trouble-the second where none were present. In a third list, we classified affec­tion about which we could not say anything, precise perimetric docu­ments being lacking.

Checking on these lists allows a first classification. Summation trou­ble only exists in lesions of the pregeniculate part of optical path­ways, they do not exist in the re­gions of the retrogeniculate part except when a stasis or vascular oedema is present.

RETINAL LESIONS AND CHORIO­RETINAL OEDEMA seem to be particularly , concerned with sum­mation difficulties.

Certain diseases require a more detailed commentary. It is not sufficient to use only ophthalmo­scope to follow choroidal evolu­tion. At the acute phase, perile­sional and lesional oedema is marked by soft isopter slopes and spatial summation trouble.

Only disappearance of these signs can allow us to say that the lesion has improved or healed and in particular that there is an absence of oedema. In retinal vascular process summation troubl­es are constant. They are evidently found in venous thrombosis and vascular obliterations as well as in slight vascular troubles such as capillarosis.

In ARTERIAL HYPERTENSION, they indicate forthcoming oedema before its ophthalmoscopic appear­ance and are a danger-signal. At the neuro-retinopathia oedematous stage, summations are evidently very disturbed, but are not of the same interest.

In DETACHMENT OF THE RETI­NA , summations are extremely mo­dified. At first stage, with diagnosis still doubtful, the revealing notch in isopters may be discovered with a certain surface-luminosity combina­tion only and not with another. This summation trouble is of great importance. It takes a greater signification after intervention. It still persists during the weeks following around coagulated retinal spots. Its disappearance means a sound and firm reapplication.

Summation study takes a great importance in CHRONIC GLAUCO­MA. They are disturbed in irritative glaucoma, that is to say in the variety of glaucoma comporting oedema. The first sign of field deterioration in that case is the well-known "bearing of the blind spot" of English authors.

In many cases, it is only possible to detect this sign through several surface and intensity combinations of the same isopter. One combina­tion inscribes an isopter outside of the blind spot, whereas another equivalent one may inscribe inside it. This sign may disappear with the pilocarpine test. As it is, at this precise stage, interventions have least chance to be dangerous, it is so extremely important to detect early signs of retinal damage. Moreover summative trouble is a sign of oedema as precious if not more than the angular scotomas.

IN ACUTE GLAUCOMA, summa­tions are evidently deeply disturb­ed, but one is rarely able to study it. In all cases oedema spreads all over the retina.

PAPILLARY OEDEMA evidently causes deep modifications in spatial summations. If summation difficul­ties are evident on medium and central isopters which is expressive of oedema of the posterior pole, the region around the blind spot must particularly be explored. In fact, curves defining the outline of the spot arise gradually one over another according to the intensity and surface of the applied test. It is thus possible to detect summa­tions. They are deeply disturbed. This trouble evidently has little diagnostical value when oedema is seen with an ophthalmoscope, On the other hand, like every functional sign, it may precede by far the pathological anatomical aspect. Stasis, for instance can often be foreseen before it actually really appears.

Later, we will see that any homonymous hemianopsia with summation trouble around the spot is caused by tumour or a vascular process, photometric disharmony of which gives a peripheric signature.

Summative sign clarifies numer­ous diagnosis. Its presence strikes out false papillary oedemas (myelin fibres, papillary drusen, hypermetr­opic pseudo-neuritis).

Unfortunately, it does not allow a diagnosis between papillitis and stasis. It must be obtained peri­metrically on other signs.

Spatial summation study at the retinal level is thus dominated by an oedema notion. It is certainly the factor in modifications of threshold sensibilities. Affections not creat­ing summative troubles are the ones which do not comport any oedema (choroidal sarcoma, non­-evolutive choroidosis, pigmentary retinitis etc.).

Leaving retinal domain to go up along optical pathways, summative trouble is less definite and less con­stant. It is detected in optic nerve and chiasmatic lesions. It is found especially in inflammatory lesions and not often in compressions.

It is then difficult to be positive but it seems again to be in relation with oedematous lesions.

Higher up, above corpus geni­culatum , no summative trouble can be found whatever be the lesion. Photometric disharmony is always an indication of peripheric lesion and when found at retro-geniculate level, it implies stasis at the first stage or a vascular process traduc­ed on retinal plane. With homonymous hemianopsia, it is an indication of tumour, with stasis threatening, or with vascular pro­cess at retinal level.

To end this survey, it may then be said that theoretical laboratory work has given perimetric clinical work a most precious sign, but, on the other hand, clinical work com­pletes laboratory research. In fact if there is no summation beyond corpus geniculatum, it is because it is of retinal origin before any­thing, though there may be a weak indication of its presence in corpus geniculatum but there is none in pathways belonging to the occipital cortex in which there must not be any convergence or influx reduc­tion. This is no proof, however that it does not exist in other path­ways of occipital associations.

In pathology, summation trouble seems to be able to be easily in­terpreted. It is mostly found with retinal oedema. Distended by im­bibition water, retinal sensory cells are stretched. They are less numerous by surface unity. Then, summation does not occur as easily. But it remains to be explain­ed why summations can be disturb­ed in lesions of the conduction ways up to corpus geniculatum, and it remains to find out which is the modification of the summation ex­ponant in various pathological pro­cesses. We have undertaken this research the results of which will be brought out later and which, al­ready keeps in store riddles of most intricate interpretation. It must not be forgotten that it is not absolute threshold but a variety of dif­ferential thresholds which is studied in clinical work, which may well deprive of all legitimately mathe­matical explanations arrived at up­to this time.

We mentioned them because actual conception is based on them and they have allowed the elabora­tion of the two variable system, but they very likely will be widely dis­cussed in a near future.

Whatever happens, the practical use of a two variable perimetry in­stead of one has proved extremely rewarding. The first great acquisi­tion can be summed up in a lapid­ary formula. The trouble of laws ruling relations between luminous intensity and surface tests belongs to the infrageniculated peripheric level of optical pathways and is not found at retrogeniculate levels.

It is especially found with oedema. These considerations find their application in the prognosis and diagnosis of many troubles and yet they derive from theoretical, mathematical, physical and physio­logical considerations, intricat­ed enough. They are a definite demonstration that theoretical re­search is not in opposition with clinical work but completes it and helps it to progress.

  Cases Where the Summation Law is Disturbed Top



Hypertensive uveitis

Choroidal wounds

Evolutive myopic choroidosis

Detachment of retina

Atypical forms of pigmentary re­tinitis

Vascular diseases of the retina

Arterial obliterations

Venous thrombosis


Hypertensive retinitis

Von Hippel disease

Chronic glaucoma

Vascular glaucoma

Subacute glaucoma

Acute glaucoma

Papillary stasis


Disseminated sclerosis

Sinusitis diseases

Tabetic optic atrophy

Optic nerve compression

Chiasmatic diseases with oedema


Opto-chiasmatical arachnoiditis


  Cases Where the Summation Law is not Disturbed Top

Choroidal melanoma

Myopic choroidosis

Pigmentary retinitis

Diabetic retinitis

Some chronic non-evolutive glau­comas

Myelin fibers

Colobomatous hole in papilla

Papillary Drusen

Toxic amblyopias :- alcohol-nicotin


- C. S. 2.


Optic nerve trauma

Optic nerve compression without stasis

Chiasmatic diseases without oedema Chiasmatic gliomas

Homonymous hemianopsias without stasis and without vascular retinal disturbances

Temporal lobe tumours without stasis

Parietal lobe tumours without stasis

Occipital lobe tumours without stasis

Retrogeniculate lesions without stasis or retinal oedema.

  Case Where the State of Sum­mation Law is Unknown Top


Oguchi's disease


Atrophia gyrata


Visceral hemorrhages with visual field disturbances

Without tension glaucoma : Von Graefe's disease

Toxic amblyopias :- lead

-methyl alcohol


-thallium acetate

-arsenic etc.

Optical neuromyelitis

Leber's disease

Optical tracts lesions

Corpus geniculatum lesions

Visual field disturbances in squint


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Spatial Summation
Physiological Sp...
Geniculate Bodie...
Cases Where the ...
Cases Where the ...
Case Where the S...

 Article Access Statistics
    PDF Downloaded0    
    Comments [Add]    

Recommend this journal