TH 7791 OQRNELL L.N1 VERSIFY IliHACA, N. ' ' 148J3 AitsVbxaty SaxteyHall PLANNING FOR DAYLIGHT AND SUNIIIGjmN BUILDINGS.* BY L. B. MARKS AND J. E. WOODWBLI,. Synopsis: The broad problems that come up in planning for daylight and sunlight in buildings are considered and the state of the art of daylight planning is reviewed. Emphasis i% l^id oif fe^Jtygienic value of daylight and sunlight in rooms, and on city planning^for good daylighting. The factors that enter into the solution of the problem of providing for adequate and suitable daylighting facilities are discussed and formula are given for the calculation of daylight illumination in buildings. By the simple expedient of a wire frame (representing the solar path) mounted on a small cardboard model of a building made to scale, the penetration of sunlight and obscuration by shadows are quickly predetermined in actual cases. A new instrument for sunlight and shadow determinations is described and illustrated. Daylight illumination meas- urements in a test room and in several court rooms of the County Court House, New York, are given. The application of the principles involved is illustrated by a discussion of the daylighting facilities of the new New York Court House, accompanied by data, charts and plans of the build- ing and court rooms. A bibliography is appended. Thi purpose of this paper is to discuss some of the broad problc-is t1^at come up for consideration in planning for day- light and sunlight in buildings, and to review briefly some of the steps that have- been taken in advancing scientific daylighting of interiors. In illustrating the application of principles involved reference will be made to a recent investigation by the writers of the natural lighting of the court rooms of the proposed new County Court House, New York City. To the lay mind planning for good daylight in buildings means simply providing plenty of window openings, just as planning for good artificial light means to the novice usually nothing more tjaan providing sufficiently powerful lamps to give a blaze of light at night. The Transactions of the Illuminating Engineering Society for the past eight years give evidence that planning for good artificial lighting of interiors is by no means a simple prob- lem and, on the contrary, involves considerations which often * A paper read at the eighth annual convention of the Illuminating Engineering Society, Cleveland, C, September 21-24, 1914. . The Illuminating Engineering Society is not responsible for the statements or opinions advanced by contributors. 2 MARKS AND WOODWBLL : DAYLIGHT IN BUILDINGS baffle even the expert. The mass of data which the Society has published has gone far to give illuminating engineering a definite status especially with reference to the use of artificial light, but thus far has. left almost untouched the vital problems connected with planning for the best use of natural light. As civilization advances we are becoming more and more a nocturnal nation dependent upon artificial light, but this is all the more reason why it t)ecomes increasingly necessary to take the fullest advantage of the benefits to be derived from the use of daylight and sunlight. In planning for the lighting of buildings our first thought is naturally to provide for illumination that will permit of good vision. But there are other considerations of fundamental and vital importance from a hygienic and pathogenic aspect. Dr. S. A. Knopf, professor at the New York Post-Graduate Medical School and Hospital, in a recent statement submitted to the Heights of Buildings Commission,^ New York City, laid great stress on the importance of securing adequate daylight and sun- light in buildings to prevent the spread of tuberculosis ; he states : Tuberculosis, which is propagated by bad air, foul air and lack of sunUght causes annually a loss of 200,000 citizens to the United States. This disease could be largely prevented would we live and work in pure air, in air relatively free from mineral and vegetable dust, and last, but not least, would we construct the buildings in which we live and labor so as to allow sunlight to enter more freely. At a recent meeting of the Municipal Art Society a prominent speaker said: I read in my newspaper to-day of the benevolent project to build a great hospital for consumptives, the victims of tuberculosis, where they may have air and sunlight. And in the same paper I read of plans for a 30-story building. What are we trying to do? What do we mean by putting up these horrible structures, to the lower floors of which no light can ever penetrate? We build hospitals for the poor consumptive, and then we turn around and erect sky-scraping structures where consump- tion may breed, so that we shall not lack for patients. In areas wjiere high buildings are crowded together most of the rooms even on the street front are inadequately lighted and 1 Report of the Heights of Buildings Commission, New York City, December 23, 1913. This report contains a mass of valuable data relating to building regulations, etc., and has been freely quoted herein. MARKS AND WOODWBLI, : DAYLIGHT IN BUILDINGS 3 many are decidedly dark. In the lower section of New York City where the office buildings range from ten to twenty-two stories high, it was found by the Heights of Buildings Commis- sion that on a bright day at noon in mid-summer artificial light was being used next to the windows in almost all of the street rooms. The conditions in the interior courts in parts of the tall building district are even worse. Fig. i gives an illustration of the conditions found in this district. The black windows indicate where artificial light was being used near the windows at noon on a bright summer day. BUILDING CONDITIONS AND REGULATIONS. Regulations Based on Street Width. — In planning for the arti- ficial lighting of buildings we locate our light sources inside of the buildings and therefore need not consider whether the build- ings are high or low, close together or far apart ; but in planning for daylighting, especially in cities, the height of buildings, their shape and distance apart are naturally fundamental considera- tions ; hence many municipalities have passed regulations restrict- ing the height of buildings and specifying the minimum street width. For example, in New York City the height of tenements is at present limited to Ij4 times the width of the widest abutting street. In the second class cities of Massachusetts, no tenement may have more than one legally habitable floor for each full lo ft. (3.0 m.) of street width, unless it is set back from the street a distance equal to the excess of its height over that per- mitted at the street line. The height of other buildings is limited to 125 ft. (38.1 m.). In Chicago, the height of tenements is limited to Ij4 times the street width while the height of other buildings is limited to 200 ft. (60.9 m.). In Boston, the height of all buildings is limited with the exception of coal hoists, grain elevators and sugar refineries. , In European cities the limitation based on street width is in most cases the fundamental restriction on the height of buildings. In America this restriction at present plays a much less important part. It may be said to be fundamental in the general height restrictions of Washington and to be of great practical impor- 4 MARKS AND WOODWE)!,!, : DAYWGHT IN BUILDINGS tance in the buildings of Boston. It is also a very important factor in height restrictions for tenement houses. In Boston, except as above noted, Charleston, Cleveland, Erie, Fort Wayne, New Orleans and Youngstown, the height of build- ings is restricted to 214 times the width of the street. With Washington, these cities are the only cities in America that base the general height limitation of all buildings on the street width. Washington is the only city that bases the permissible height of buildings upon the width of the street diminished by an arbitrary amount; the height of buildings on residential streets more than 70 ft., (21.3 m.) in width may not exceed the width of the street diminished by 10 ft. (3.0 m.). In cities of the second class (population 50,000 to 175,000) in New York State, no building to be used for living purposes, except a hotel, may exceed in height the street width nor in any case may exceed 100 ft. (30.5 m.) in height. Regulations Based on Maintenance of a Minimum Angle of Light. — A limitation based directly on street width maintains a constant minimum angle of light for the front of the building at the ground floor. If the prescribed height is equal to street width this minimum angle of light is 45 deg. ; if i J4 times street width, it is 33^ deg. ; if 2 times street width, it is 26j4 deg. ; if 2}4 times street width, it is 21% deg. The converse of this is that the maximum angle of light obstruction will be the difference between the above amounts and 90 deg., i. e., 45 deg. for height limit equal to street width; 56 J^ deg. for height limit i^/^ times street width; 63 J4 deg. for height limit 2 times street width; 68j^ deg. for height limit 23^ times the street width. From Fig. 4 it is clear that a flat limit of height is not necessary in order to secure a minimum angle of light. If the height limit based on street width is made to apply only to the elevation of the building at the street line and other portions of the building are set back in the same ratio as height limit to street width the angle of light is maintained. If the height limit is twice the width of the street a set-back after reaching the height limit at the street line of 5 ft. (1.5 m.) for every 10 ft. (3.0 m.) of increase in height will maintain the angle of light at 26J4 deg. In London, the angle of light is more expressly stated. The Fig. J..— Use of artificial light in offices on Exchange Street, New York City. 'I'he black windows indicate where artificial light was being used near ihe windows at noon on a sunny suninier day. Fig 2— Cardboard model of Xe\\- York Court Hou>?e made to scale showiiia wire frame I representing the solar path) mounted to determine penetration of sunlight and obscuration by shadows. ^f Mf^M^ H=c: ■is- 3— Archittcls' typical court room plan, No.;,, new County Court House, New York City, .showing by shading the grouping of court rooms into separate unils. MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS 7 rear heights of a building are generally regulated by a line drawn at an angle of 63^^ deg. to the horizontal toward the building from the rear line of the lot. That is, a building may not be built so as to obstruct the light of the adjoining lot in the rear at an angle of more than 63^^ deg. Height Limitations in American and European Cities. — The maximum height limit in America is, as a rule, set so high that it limits the height of buildings only when what might be termed the logical height limit for that particular city or locality has been very much exceeded. In other words, the maximum height limit is no height limit at all so far as most buildings are con- ANGLt OF LIGHT llj* ■ I • OBSTRUCTION 68f AN6LE OF LIGHT t6t' ■ 0B5TR.63i ANGLE OF LIGHT 33i- ■ OBSTR.SSi" AN GLt OF LIGHT 45" -OBSTR45' FACTORS OF STREETWIOTH » SIRCETWIJTH Pig. 4. — Angles of light and of light obstruction ou ground floor on street front. cerned ; it only prevents the erection of a few exceptionally high buildings. It does not limit or condition the character of the great mass of buildings erected even in the central business dis- trict. In Boston, Chicago, and Washington, however, the present maximum height limits do constitute a very practical restriction, as evidenced by the tendency in certain districts to build up to the full height allowed by the restrictions. A tabulation of height limits in some American and European cities is given in Appendix I. DAYLIGHT. Determination of Sky Obstructions. — In most cases, planning 8 MARKS AND WOODWELI, : DAYLIGHT IN BUILDINGS for daylight involves consideration of sky obstructions, such as buildings or other structures, or natural objects such as an ad- jacent hill or mountain, etc. The interception of daylight by such sky obstructions depends upon several factors, the most important of which is the average height of the sky obstruction and its position upon the horizon with reference to the room or building. For example, a tall adjacent building situated off to one side within the line of light intercepted by the window apertures will have little influence in reducing the light entering the room, whereas if the same building were situated directly opposite the window of the room, it would result in cutting down a large proportion of the daylight that would otherwise enter the room. Similarly, a long line of low buildings or distant buildings would have a very much less effect in diminishing the light entering the room, than would a corresponding area of sky obstruction situated at such a point as to intercept the light that would other- wise enter the room at higher and more favorable angles. 1. Photographic Method. — A sky obstruction forms an artificial sky line which is above the natural horizon; it represents the outline of buildings and ground or other obstructions. This sky line can be approximately determined by taking a series of over-lapping photographs of the horizon from the same station, the series to include such portion of the entire horizon as it is desired to show. By properly joining, a panoramic view is se- cured showing the actual sky line formed by the obstructions. The natural horizon (which will be a horizontal line drawn across the center of the picture if the camera is held level) is first de- termined on the panorama, and from the known angle of the camera lens (corrected for distortion) parallel lines representing the degrees of altitude above the natural horizon can be drawn on the panorama, from which the altitude of the artificial ob- struction may be determined at any point of the compass. 2. Mathematical Method. — Another method for determining the artificial sky line consists in measuring or determining from a map the distance from the point at which it is desired tovde- termine the results, to given obstructions and at the same time using the known height of each obstruction from which the altitude in degrees of each obstruction can be readily figured MARKS AND WOODWBLL : DAYLIGHT IN BUILDINGS 9 and an artificial sky line plotted. Except in the simple cases this method involves much more work than the photographic method. The object of such a panorama is to determine the percentage of sky obstructed in a given direction, and this may be deter- mined with sufficient accuracy by drawing upon the photographic panorama or graphically constructed sky line, horizontal lines representing the degrees of altitude (5 deg. intervals are gen- erally sufficient). The approximate area of sky obstructed in each of the zones of altitude may then be easily determined by inspection or, if more accurate results are required, by means of a planimeter. In view of the fact that the panorama is drawn with horizontal and vertical co-ordinates, and is on a plane surface instead of being upon the surface of a sphere representing the visible sky, it is necessary to multiply the area of obstruction in each separate zone of altitude by a factor which will give the proper weight to each of the horizontal zones. The approximate factors or relative areas for each 5 deg. zone from the horizon to zenith are given in the accompanying table. Typical sketches showing sky lines determined in accordance with the above methods are shown in Appendices 4 and 5. Table giving relative area of zones at 5 deg. intervals from horizon to zenith ; horizon of 180 deg. with radius unity: Zone: degrees above Factors or .-.. the horizon relative areas o deg. to 5 deg. o. 2738 5 " " 10 " 0.2716 10 " "15 " ".2675 15 " "23 " 0.2613 20 " "25 " 0.2532 25 " "30 " "-2432 30 " "35 " 02310 35 " "40 " 0.2173 40 " " /?5 " 0.2020 45 " "50 " ,• o 1S55 50 " "55 " U.1670 55 " "6° " "-1475 60 " "65 " 01 264 65 " "70 " o 1048 70 " "75 " 00824 75 " "80 " 0.0593 80 " " 8.=i " 00358 85 " "go " 0.0120 Total area of quarter sphere representing the sky visible between horizon and zenith with horizon of 180 deg- • 3. 1416 lO MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS It is also possible to determine from such a panorama the average angle of sky obstruction, and the consideration of this angle is an important factor in planning for daylight, as it is obvious that a sky obstruction at a very low angle will be very much less efifective than a sky obstruction at a higher angle, and also that the depth of the room and the angle which may be considered the most favorable for the entrance of light in a given room will have an important bearing in determining the reduction caused by the sky obstruction. The effect of these sky obstruc- tions will also depend to a considerable extent upon the color and reflecting power of the obstructions which in the case of buildings may modify the effect materially. For example, a tall building of light color may return a great deal (often more than 50 per cent.) of the light received by it, and present a surface which may have a brightness as great as if not greater than that of the sky. In taking account of the effect of sky obstruction in planning for daylight in a given case, it is necessary to consider not only the existing factors and conditions so far as they may be deter- mined, but also the possibility of future changes in buildings, etc. The four general cases which are encountered are : 1. Unobstructed sky ; 2. Partly obstructed sky — builditigs of low diffusion ; 3. Partly obstructed sky — great diffusion; 4. Sky completely obstructed. To determine comparative results quantitatively for the four cases, the most practical way is to make tests by a luminometer under conditions which most closely resemble those of the prob- lem in hand ; such tests must of course be made under conditions which are most typical and representative of the working condi- tions under which the daylight is to be used in the building or rooms for which its utilization is being planned. The variations due to the seasons of the year in the presence or absence of snow on the ground, and the orientation . of the building with reference to the points of the compass, etc., etc., should be given due weight. Illumination Due to Direct Light from the Sky on the Vertical Surface of a Window of a Building Front, with Obstructed Uniform Horizon.— In Fig. 5, let A and B represent the buildings MARKS AND WOODWELI< : DAYLIGHT IN BUILDINGS II on the opposite side of the street of width W. The formula for determining the illumination on the vertical window at point of observation a in building A in a case where the building B oppo- site is of constant elevation H above the point of observation a is derived as follows : The sky may be represented by a spherical surface with its center at the point of observation a. A plane passing through the point a and through the top line of the opposite building front cuts a circle from the surface of the sphere drawn to represent the sky, and the opposite building obscures all the direct light from the sky which would otherwise enter below the plane drawn through a and the top line of the front of building B, and the Fig. 5.— Illustrating the derivation of the formula for determining the illumination due to direct skylight falling on the vertical surface of a window of a building front with obstructed uniform horizon. amount of sky obstructed may be represented graphically by the projected intersection of this plane with the spherical surface drawn to represent the sky, giving the semi-ellipse e-a-f-d-e. The projected area of the visible sky is therefore the lune f-c-e-d-f. The area of the semi-ellipse is the area of the semi- circle on the sphere projected on the plane through a and c as shown by the semi-ellipse and is equal to ; >4xR^ sin h where h is the angle between the line ad and the horizontal. Sub- tracting this area from that of the semi-circle e-c-f , J^ttR-. gives : J^7rR^(i — sin/t) for the area of the lune. If B represents the brightness of the sky in candles per square foot, the illumination at a in foot-candles will be: 12 MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS I = )^ff B (i — sin A). If we use W as the width of the street, that is, the distance from A to the face of the building, and H as the height of the opposite building front, not above the ground, but above the point of observation, a, vvre can express sin h in terms of H and W, thus : H sin h = 1 H'^ -i- W^ and our expression for the illumination at a becomes : I- J^ttB (i -— =^^) (I). Illumination on a Horisontal Plane within a Room. — The illumination directly due to the sky at a point on a horizontal plane in a room depends on, (a) the amount of sky visible from that point, (&) the brightness of this area of sky, and (c) its anguar elevation. If the light comes through vertical windows, the solid angle subtended by the visible sky at any working posi- tion in the plane of reference is usually small enough for the illumination in foot-candles there to be given by the formula: I = B (osin 61 (2). Where B, represents the brightness of the sky in candles per sq.ft. o), the solid angle subtended by the visible sky at the test station. 6, average angle of elevation. The solid angle may be expressed in "square degrees" as used by Weber and Cohn, and more recently in the British committee report on daylight illumination of schools in which the recom- mendation is made that "The darkest desk in any school room should receive an illumination equivalent to that derived directly from 50 reduced square degrees of visible sky." A square ydegree is equal to thcsolid angle subtended by a square each of whose sides subtends one angular degree. Factors Governing Daylight Admission. — The amount of day- light available from a window depends upon — (o) The size, shape, depth and position of the window opening. (b) The brightness of the sky. (c) The sky angle, that is, the angle formed by a vertical plane passing through a window in the wall of a building, and a line from the window through the top of the opposite build- ing. (In the case of an unobstructed horizon the sky angle is 90 deg.) MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS T3 (d) The diflfusion from the street surface and from adjacent buildings. (e) The character of glassware in the windows. Only a comparatively small percentage of the flux of daylight that reaches the window is effective in the illumination of the interior. The effect of blinds, curtains and other decorative and absorbing media in obstructing the entrance of light is often con- siderable. Furthermore if the color of the walls and furnishings is dark a large portion of the Hght is absorbed. Fig. 6 shows two windows of equal area, one of which is vertical and the other horizontal. These two windows will give very different results in the lighting of a room. Daylight will yig. 6.— windows of equal area. Daylight will be received through the vertical window at a much more favorable angle for interior illumination, be received through the vertical window at a much more favor- able angle for interior illumination, even with an unobstructed horizon in both cases. If the sky is obstructed for some little distance above the horizon the vertical window will have a still greater advantage over the horizontal. Obstruction to Light at Window Openings. — It is customary to determine limiting angles at which the light is completely inter- cepted by window walls, cap stones, columns and other obstruc- tions to light ; for example, in Fig. 7 the daylight received in the Court House window shown is from a horizontal visual angle of 129 deg., and from a vertical visual angle of 73 deg. ; from such 14 MARKS AND WOODWBLL : DAYLIGHT IN BUILDINGS measurements the proportion of the total amount of daylight intercepted by the windows may be determined. Relation of Window Area to Floor Area, etc. — With reference to the size of the windows for a given interior room, certain ratios are frequently used as a basis for design and comparisons. Among the ratios used are the following : Ratio ojE window to floor area. Ratio of window area to depth of room. Ratio of height of windows to depth of room. Ratio of window area to outside or window — wall area. In some classes of interiors the ratio of the area of the window space to the floor area to be lighted is sometimes specified to be E LE VAT I O N Fig. 7. — Plan and elevation of a window showing obstructions to admission of day-, light and limiting horizontal and vertical visual angles. I to 4 or I to 5, as in factories for example; while in other cases such as office buildings, this ratio may be specified as I to 7 or I to 10, etc., depending to some extent upon the depth of the room. This system of computation, however, is evidently limited in its application, because the ratio of window space to floor area taken alone does not give a true relative value of daylighting facilities. For example, the ratio of window space to floor area for the lower floors of a building might be quite different than that of the upper floors for equivalent illumination, especially if MARKS AND WOODWEtly : DAYWGHT IN JSUILDINGS I5 part of the direct light of the sky is cut ofif by an adjacent tall building. While the ratios above referred to are often of considerable value in comparing the window openings of rooms of the same character, arrangement and exposure, the relative values indi- cated may be very misleading, as for example where the shape of a room is quite put of the ordinary or where windows of unusual shape are provided. Limiting Angle of Light. — Account is frequently taken also of the angle of a point half way between the top and bottom of the window and a given reference point within the room, such as a point 2 ft. 6 in. (0.76 m.) or 3 ft. (0.9 m.) above the floor at a working distance near the most distant wall; account is also taken of the angle between such a reference point and the top of the window. Light-value of a Window. — It has already been pointed out that the ratio of the window area to the floor area taken alone may convey a very imperfect and even erroneous idea of the amount of daylight that reaches an interior. This is clearly shown in Fig. 6 previously referred to, in which two windows of the same area but differently located with respect to floor space, obviously give very different results in lighting. The ratio referred to does not take cognizance of the effect of obstructions outside of a window, of the amount of sky visible, of the contour and depth of the window or of the height and depth of the room. Prof. L. Weber of Kiel, Germany, introduced the conception of the "light-value" of a window to give a more correct basis for estimating daylight entering a window. The "light-value" is the ratio between the actual illumination of the vertical window pane and the illumination which would be received from an unob- structed sky; that is to say, the ratio between the solid angle subtended at the window by the visible sky to the solid angle subtended by the sky with free horizon. If this quantity is modified so as to include the percentage of the window area through which unobstructed sky can be seen it becomes even more useful. If the light value be denoted by L and this be multiplied by the permissible ratio of window space to floor area we obtain a new quantity, P, which is a measure of l6 MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS the usefulness of a window for admitting daylight. Having determined L,, we have F — L,.~- where £■ is the area of the glass surface of the window and b the area of the floor of the room. This formula is of course limited in its application as ex- plained in the preceding paragraph. Ratio of Indoor to Outdoor Daylight. — Cohn found that a school room was sufficiently lighted by day if the minimum hori- zontal intensity on any desk was about 2j4 foot-candles. Weber found that with the average value of the sky brightness likely to be met during the hours of study, this illumination should be obtained on any desk in a school room receiving 0.5 per cent, of the unobstructed daylight illumination out of doors; that is to say, 0.5 per cent, of unobstructed roof light. As the vertical window is exposed to only one half of the sky with a free horizon, the figure given by Weber is equivalent to i per cent, of the daylight on the window sill. Window Glass. — In cases where buildings are necessarily built close together or where it is impracticable to secure a reasonably unobstructed horizon, it is possible to redirect the light from' the sky by means of prismed windows. By this means the light in the rear of a room say 60 ft. (18.3 m.) deep may easily be increased to from 10 to 15 times, though it must be understood that the total flux of light in -the room is not increased by the substitution of prismed windows for clear glass. When windows are equipped with prismatic glass the incident rays of daylight are deflected from the course which they would naturally take if plain glass windows were used and instead of striking the floor at points near the position of entry of the light, the rays reach the inner portions of the room thus lighting parts of the working section where more light is required. In the crowded tenement districts in large cities the substitu- tion of daylight prisms for ordinary glass would be a boon to those who are now compelled to live in rooms that scarcely see the light of day. Fortunately the movement now on fopt to enforce regulations providing for direct daylight and sunlight in tenement houses will soon make the dingy tenement a thing of the past. II ,2 SUNRISE SUNSET SUMMER SOLSTICE II 12 SUNRISE E EQUINOXES ,0 II iz WINTER SOLSTICE Fig. 8.— Apparent path of the sun at latitude of New YorkYity illustrating angles of sun light at different hours of the day at the solstices ayd equinoxes. ,:Vj ^Sy DIRECTION OF sun's BAYS ; 1 ■ ^S^''' NOON, WINTER SOLSTICE ' '% * g F B'^ ^^I^^^B^ • JkU-ttUOJUU HII^^'''Jiiiilwt^-'^i^ ' ^"Zv' /■V:n:^;::,.. i^^ •■ ; Fig. q— Plaster incdel of new County Court House, New York City, shoM-ing wire frame (representing the solar path) mounted to determine direction of sun's rays for all seasons of the year. MARKS AND WOODWEtL : DAYUGHT IN BUILDINGS I9 SUNLIGHT AND SHADOWS. Variations Due to Latitude, Seasons of the Year and Hours of the Day. — The apparent path of the sun at the summer and winter solstices and the equinoxes for latitude about 41 deg. ' north (approximately that of New York City) is shown in the perspective diagrams, Fig. 8. These diagrams also show the angles of sunlight at the different hours of the day, solar time. In order to determine in a practical way the direction of the rays of the sun at any hour of the day for the winter and sum- mer solstices and the equinoxes, a simple frame representing the celestial sphere was constructed of wire with the apparent sun paths shown by wire circles as indicated in Fig. 2. The com- mon axis of the circles representing the apparent paths of the sun makes an angle with the horizon corresponding to the lati- tude chosen, namely 41 deg. in this case. Hour marks (con- sisting simply of pieces of twine, knotted) were placed at 15 deg. intervals on each of the three circles. Another circle of wire was placed on the north and south axis of the wire frame, and thfe points of intersection of this circle with the circles representing the apparent paths of the sun, correspond with the zenith or twelve o'clock position of the sun, solar time. By placing an artificial light source of small size at some distance from the wire frame the direction of the rays of the sun could be reproduced for any hour of the day, or for any of the four seasons of the year by locating the artificial light source at such a point that the rays of light are in line with the desired hour of the day on the sun path circle representing the season chosen and the center of the wire frame corresponding with the center of the celestial sphere. The wire frame may then be mounted on a miniature wood or cardboard scale model of a building or buildings adjusted to correspond with the points of the com- pass established by the wire frame. By using models of light color, a partially darkened room and a concentrated beam of light from a miniatui"e lamp with parabolic reflector, it is possi- ble to secure clear, sharp shadows on the scale models, suf- ficiently accurate to form the basis of conclusions with respect to the penetration of sunlight and the obscuration by shadows in any given case. 20 MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS From the observations thus taken drawings or charts may be made showing the direction of the sun and the shadows pro- duced during any period of the year and for various hours of the day between sunrise and sunset. Fig. 2 shows the wire frame set up on a small model made to ' scale for the purpose of determining the shadows cast by the inner court and connecting bridges on the inner wall of the pro- posed new County Court House, New York City. Typical shadows thus determined are shown in Appendix No. 6. The degree of penetration of sunlight in the court rooms during the working year was determined by this method. The use of this wire frame for determining the direction of the rays of the sun on the large scale model of the court house is shown in Fig. 9. Some idea of the great change in direction of light due to the change in the path of the sun from the winter solstice to the summer solstice, and the consequent effect upon the production of the shadows at different times of the year may be obtained by reference to this figure. Molesworth's Planisphere Diagram. — Molesworth^ describes a unique, graphical method whereby with the use of the plani- sphere, — or the plane projection of a sphere, — the hours during .which any particular window receives direct sunlight can be de- termined. This method is mathematically correct, but requires the following data with regard to each obstruction, namely: — (o) The angular elevation of the different salient points. (6) The true bearing of these points, (c) The true bearing of the side of the building in which the window is located, (rf) The latitude of the place. As this information is often difficult to obtain, and mus^ be secured for each important obstruction on the visible horizon, considerable time is required to obtain a definite result, even for one season of the year and the process must be repeated if the information is required also for other times of the year. The determination of shadows by means of models is much simpler than that by the use of descriptive geometry or calcula- tion, or by the graphical method of using a planisphere diagram. Moreover the models may be readily shifted to study the effect - Molesworth, H. B.; Obstruction to Light; London, 1902. o ■o I d in Fig. II. — Cardboard models of buildings made to scale showing shadows at noon hour, (i) summer solstice, (2) winter solstice, (3) vernal and autumnal equinoxes. MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS 23 of orientation with respect to the points of the compass, and the effect of the shadows at various times of the day may also be easily compared and studied, as well as the changes due to the varia- tions of the path of the sun at the various seasons of the year. Even in the case of buildings of complicated shape, the shadows may be determined with sufficient exactness for all practical purposes for any time of the day and any time of the year and for any position of the building. A New Instrument for Sunlight and Shadow Determination. — A new instrument designed by J. E. Woodwell for rapidly and accurately determining the apparent position of the sun at any hour of any day in the year is shown in two views in Fig. 10. This instrument is constructed in accordance with the same un- derlying principles as those that hold for the wire frame prev- iously described, but it is adjustable for any latitude between the equator and 60 deg. north or south latitude, and also for any time of the year. The instrument consists of a stationary base A carrying a horizontal or horizon circle B to set the instrument at any de- sired azimuth. The vertical post C carries a vertical support for the vertical circle D by which adjustments are made for any degree of latitude from the equator or zero to 60 deg. north or south latitude. Upon this adjustable vertical circle D a station- ary circle E is so mounted that the axis of the latter passes through the axis of the vertical circle and the vertical axis of the horizontal or azimuth circle. This circle E serves as an' hour circle, each hour being represented by 15 deg. intervals. Settings for the different hours of the day are made by means of a revolving arm F. The revolving arm F carries with it a support for the adjustable sliding arm G, which has, at its inner end a small circular aperture through which a beam of light may be projected upon the crosshairs so mounted upon an adjustable support H that the center of the crosshairs lies at all times in the vertical axis of the horizon or azimuth circle B, the center of the vertical circle D and in the axis of the hour circle E. The adjustable sliding arm G is set perpendicular to the plane of the hour circle and is calibrated to correspond with the set- 24 MARKS AND WOODWBLI/ : DAYLIGHT IN BUILDINGS tings of the instrument for any day of the year. At the dates of the equinoxes, for example, the sliding arm G is so set that a beam of light received from an artificial light source so placed as to correspond with the direction of the sun, would pass through the circular aperture in the sliding arm G and intersect the crosshairs at the center of the instrument in a line parallel to the plane of the hour circle and perpendicular to the axis passing through the center of the hour circle and the center crosshairs set to represent the latitude of the place for which the instrument is adjusted. Due to the inclination of the axis of the earth at 23I/2 deg. the settings for the summer and winter solstices are made by sliding the adjustable arm G towards or away from the hour circle an amount which is equivalent to the angular distance of 23J/2 deg. either side of the setting for the equinox. The set- tings for intermediate months and days of the year are made in a similar manner by reference to astronomical tables giving the declination of the sun at zenith, etc. The instrument may be levelled by means of the adjusting screws and the levels mounted on the base at right angles to each other, and when the vertical circle is set for the equator or at zero, the intersection of a beam of light passing through the second circular aperture J and the crosshairs at the center of the vertical circle will be in the plane of the horizon. A point on the horizon may be established at any point of the compass by merely revolving the azimuth plate upon which the instrument is mounted. As in the case of the wire frame previously described repre- senting the celestial sphere, the best results in determining shadows upon small scale models of buildings, etc., are secured by using a miniature lamp and a small parabolic reflector, the artificial light source being placed at a sufficient distance away from the model in proportion to the size of the latter to avoid distortion due to non-parallelism of the rays of light. In practise it has been found that a distance of ten times the greatest dimension of the model is ample to secure accurate results. This instrument may also be used to determine the form of ■shadows produced by rays of light from certain definite direc- MARKS AND WOODWELI, : DAYLIGHT IN BUILDINGS 2$ tions upon architectural moldings, cornices, relief work, and also for determining the angles at which direct reflection of light is produced in connection with problems in interior illum- ination. The accompanying photograph, Fig. ii, shows the obscuration of the direct sunlight at different times of the year for a specific group of buildings by means of small card-board models of the buildings constructed to scale, (i) shows the obscuration of the direct sunlight and consequent shadows at noon at the summer solstice. (2) shows the shadows for the same group of buildings at noon at the winter solstice. (3) shows the shadows of the same group of buildings at the vernal and autumnal equinoxes. The shadows for any intermediate period may be similarly shown. DAYLIGHTING FACILITIES OF NEW COURT HOUSE. The new County Court House to be erected in New York City at an estimated cost of $10,000,000, presents unusual problems in daylighting owing to the architectural features of the building which was designed by Mr. Guy L,owell, architect. The plans of the Court House show a circular building approxi- mately 400 ft. (121.8 m.) in diameter and 200 ft. (60.9 m.) high containing an annular court or open space iij ring form 40 ft. (12.2 m.) wide. This court is crossed by bridges connecting with the main building. The inner wall of this court rises 75 ft. (22.8 m.) above the floor line of the fourth floor. The following preliminary plans of the building are submitted in Fig. 3 and Appendices 7, 8 and 9 : typical court room plan showing grouping of court rooms into separate units ; section of building on north and south axis according to architect's plan No. 3;. plan of typical court room corresponding to plan No. 3, showing windows and light court, arrangement of room and exterior columns; section of typical court rooms through fourth and fifth floors at colon- nade, corresponding to plan No. 3. Obstruction of Light by the Walls of the Annular Court. — The reduction of sky angle due to the obstruction of light by the walls of the annular court and by the connecting bridges shown in the preliminary plans submitted will result in cutting out a large percentage of the light that would otherwise reach the working portions of the court room through the rear windows on the 26 MARKS AND WOODWELI. : DAYLIGHT IN BUILDINGS fourth and fifth floors. In view of the ample provision of light through the large windows in the outer wall, this reduction is not serious with respect to adequacy of lighting facilities but rather with respect to flexibility; that is to say dependence must be placed very largely on lighting from one side (the outside windows) only, as the light coming through the windows on the court side would not alone be sufficient to give adequate illumin- ation of the room. The flexibility possible with equal lighting facilities on both sides of a room is often a great advantage, especially when the erection of high buildings close by and im- mediately opposite the windows results in obstructing the light on one side ; and also when it becomes necessary to screen direct sunlight on one side during working hours. Reduction of Light Due to Surrounding High Buildings. — From observations made on the site of the proposed new Court House at a height of loo ft. (30.5 m.) above the ground (fourth floor court room level), it was found that at the present time there is substantially no sky obstruction at this level with the exception of that due to a few buildings on the south and west, the sky angle averaging approximately only 12 deg. A chart showing the sky angle at the fourth floor level for 360 deg. is given in Appendix 5. The probable growth of tall buildings in the vicinity in the future will considerably reduce the effective daylight on the lower floors of the Court House. The Effect of the Contour and Depth of Windows and of the Exterior Colonnade upon the Visual Angle. — What is sometimes termed the visual angle of the window is the angle between two planes that pass through the edges of the glass at the sides of the windows and the furthest projecting edge of the building wall or column, as the case may be. It is very important that the exterior colonnade be located close to the outer wall of the building as a very slight increase in the distance of the columns from the window wall results in a material decrease in the visual angle of the window, as may be readily seen from the diagram Fig. 7. Calculations show that the increase in the visual angle that would be obtained by the removal of the columns from the present location shown on the plans, would not secure a material MARKS AND WOODWBLL : DAYLIGHT IN BUILDINGS 27 increase in effective daylight within the rooms. Hence from the lighting standpoint there is no need of sacrificing the exterior colonnade which may be said to constitute an indispensable ele- ment of beauty of the exterior. The contour and depth of the windows as revealed in the plans are favorable to good lighting. On the fourth floor the pro- jecting stone above the lower window will considerably obstruct the ingress of light through this window. However, since the main lighting of the room is carried out by the upper windows this obstruction will not reduce the total daylight in the room to any material extent. Window Area and Visual Angle. — The ratio of the floor area to the glass area, and the visual angle taken 3 ft. (0.9 m.) above the floor at the wall opposite the windows in several court rooms in the present New York County Court House and in the princi- pal court rooms of the new Court House are given in Appendix 10. Daylight Illumination Measurements. — Measurements of day- light illumination were made in a test room to determine the actual intensity of daylight on the working plane at different times of day and on different days. Similar measurements were made in several court rooms of the present Court House. The rooms chosen are a court room said to be favored by several of the justices because it has windows on three sides, thus providing for better lighting and ventilation than in most of the other rooms ; a court room having three windows on one side only ; and a court room having two windows on one side only, chosen because both windows were exposed to direct sunlight. In the last of these, measurements were made when direct sunlight was shining into the room and also when the windows were screened by amber colored shades. The charts submitted con- tain the records of illumination tests in all these rooms (see Ap- pendices 2 and 3). Unilateral Lighting as Compared with Lighting from Two or More Sides. — In planning the lighting of school rooms there is a strong tendency towards the adoption of unilateral lighting, with windows located in the wall to the left of the pupils. When this plan is adopted it is desirable that the ceiling, and wall op- 28 MARKS AND WOODWBLL : DAYLIGHT IN BUII,DINGS posite the windows should be light in color to secure suitable diffusion of daylight. Calculations from the plans submitted for the Court House show that more than 90 per cent, of the light on the working spaces of the fourth floor court rooms will come from the three windows in the outside wall, as the two windows in the wall looking out on the annular court will receive very little direct light. On this floor therefore, if the present plan were followed, we would have a close approach to unilateral lighting. In fact the two windows in the court wall would prob- ably not contribute as much to the effective lighting of the main portion of the room as would a wall placed in front of the wing to be used for a visitors' gallery. Such a wall (light in color) would considerably increase the diffused light in the room. The chief advantage of the two windows on the court side of each room on the fourth floor would be from the standpoint of ventilation rather than that of lighting. On the fifth floor more than 80 per cent, of the light on the working spaces would come from the three windows in the out- side wall according to the present plan. The typical court room shown on the plan has a wing contain- ing the visitors' raised platform. The disposition of windows in this room is such as to secure adequate daylight and satisfactory ditribution of light, provided the walls and ceiling are made light in color and suitable means are employed for screening direct sunlight without darkening the room to a prohibitive degree. Although such means can be readily devised it is considered de-, sirable to insure greater flexibility of lighting facilities on the fourth and fifth floors, by added lighting from the court side so as to permit of satisfactorily lighting the room mainly from either one side or the other, thus securing substantially all of the advan- tages of unilateral and of bilateral lighting. Lighting from more than two sides is almost always open to serious objections. In a city like NeW York, especially in lower Manhattan where high office buildings are likely to be built close together, the pro- vision of a surplus of daylight from the open court of a large building would have the added advantage of compensating for MAKKS AND WOODWELI, : DAYLIGHT IN BUIIvDINGS 29 Y obstruction of light by high buildings that may be erected in ; vicinity in the future. There is a great diversity of opinion as to the relative merits of ilateral, bilateral, and skylight lighting for different classes of eriors; except in the lighting of school rooms very little pro- :ss has been made towards standardization of methods of day- hting. Effect of Color of Walls. — The color of walls and ceiling and of room furnishings plays an important part in the distribu- n of daylight within the interior. A window area that would adequate for the provision of good daylight within the room the walls and ceiling are light in color, might be inadequate if ; walls are dark in color. With unilateral lighting or with hting that comes principally through windows located on one le of the room a light colored interior finish to secure good fusion of light is a pre-requisite, because with dark walls there luld be a marked tendency to strong contrast of light and idow on the faces of those who do not face the windows — an jection which would have considerable weight in a court room. Effect of Screening of Windows to Exclude Sunlight during orking Hours. — All of the court rooms are provided with win- ws on both sides; hence there will never be need of screening of the windows to exclude direct sunlight. In the present plan, s flexibility of lighting facilities leaves little to be desired in ; court rooms on the sixth floor, but is not secured on the fourth 1 fifth floors. However, with the proper use of suitable shades screen the direct sunlight, good lighting conditions may be intained on all of the court room floors. \ photometric test made in one of the court rooms lighted from z side only in the present County Court House Building, Dwed that when the windows were lighted by direct sunlight 1 the shades (amber colored) drawn to exclude the sunlight the required degree, the occupants of the room still had suffi- it light to carry on their work with comfort though the intensity illumination was relatively low. The foot-candle intensities of flight illumination in the court room under these conditions ; given in Fig. 2, Appendix 3. It will be noted that on the Ige's desk the intensity of daylight illumination was 2.75 foot- 30 MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS candles (the minimum intensity in the working portions of the room). This intensity was considered quite sufficient for com- fortable reading of law books and other papers before the court. Penetration of Direct Sunlight. — A study was made of the ex- posure of the court rooms with reference to the penetration of direct sunlight at all times during the year ; the calculations were checked by tests on a model of the building scaled according to plan 3. Sample charts containing information on which the periods of direct sunshine in the court rooms may be determined are shown in Appendix 6. Based upon the preliminary design submitted, the tests and calculations show that a large percentage of the court rooms will receive direct sunlight on sunny days for at least an hour or two a day during the working term of the year. During the period from the late fall to the early spring there will be no substantial penetration of direct sunlight through the court room windows facing the court on the fourth floor, and for the same period five of the court rooms on this floor will receive no direct sunlight whatever during the working year. MARKS AND WOODWELL : DAYLIGHT IN BUII^DINGS 31 APPENDIX I. TABUI.ATION OF BUILDING HEIGHT LIMITS IN AMERICAN AND EUROPEAN CITIES. American Baltimore 175 feet Boston' District A ( Business and commercial) 125 " District B ( Residen- tial section) • . .80-Too " Buffalo^ Charleston' 1 25 " Chicago 200 ' ' Cleveland' 200 " Erie' '. 200 ' ' Indianapolis 200 " Los Angeles 150 " Manchester, N. H ■ ■ . 125 " Milwaukee 225 " New Orleans' 160 " Cities Portland, Ore 160 feet Rochester^ Scranton 125 " Youngstown' Fort Wayne' 200 " Providence 120 " Salt Lake City 125 " Toronto' 1 30 " Washington D. C. Pennsylvania Ave 160 " Business bts' ... . 130 " Residence Sts*. .. 85 " Seattle about 20 stories European Cities Aix-la-Chapelle 65.6 feet Altona 72.2 " Berlin 72.2 " Bremen 62.3 " Breslau 72.2 " Cologne 65.6 " Dortmund 65.6 " Dresden 72.2 " Duisburg 65.6 " Dusseldorf 65.6 '•' Edinburgh.... 60.0 " Elberfeld 65.6 " Frankfort 65.6 " Halle 590 " Hamburg • 78.7 " Hanover 65.6 feet Kiel 72.2 Leipzig 72.2 London 80.0 Liibeck 59.0 Magdeburg 65.6 Munich 72. 2 Paris 65.6 Posen 65.6 Rome 78.5 Stockholm 72.2 Stuttgart 65.6 Vienna 82.0 Zurich 43. u 1 Not to exceed 2j^ times width of widest street. - Not to exceed 4 times average least dimension. 3 Not to exceed 5 times least dimension at base. < Not to exceed street width plus 20 feet. 5 An intermediate height between 60 feet and 85 feet on streets over 70 feet wide- height not to exceed width of street minus 10 feet; 60 feet on streets from 60 to 70 feet wide: and street width on streets less than 60 feet wide. 32 MARKS AND WOODWELIv : DAYIvIGHT IN BUILDINGS ^^^^^ MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS 4 33 t ' — ±- >,«(— ^ ^2: ; r z\ • — -s-.n » ^Sa IxH^^iSl^WSWWVi l\\N.V>^ '3iavn3SNnoD fcSo Si' /^U|^«- ^ - u#== (e r« 1— =Tl=d iiiEp5Sn;iipESSsg| 34 MARKS AND WOODWELIv : DAYLIGHT IN BUILDINGS Hinos SrTT"! o u P. < 3Qniinvjos.93a i.SV3i HiaON of-|~t Si rr -t-r* r^T E!^ "I Hinos Sj-- I I I 1 I I I 3animvjos.33a MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS 35 D^&v-'Wao- ,-C— -i rfl'-Q^ DP ' 00 QO UD D 0cf OQ DD no JC "ac DC 'do 00 W^ffi" L^O QO or DO 00700"'' .Av 1 ;3d>(i -? •=* -^ f*? Appendix 6.— Shadows thrown on court room windows by the inner court and connecting bridges new New York County Court House, plotted for months of October and February, shadows in morning hours similar to those in afternoon. 36 MARKS AND WOODWELL : DAYLIGHT IN BUILDINGS MARKS AND WOODWBl,!, : DAYLIGHT IN BUILDINGS 37 ''®^"~^ ^7r:v7.^'-^/^:-'''.;>:^^:y'7,r/Z7^^^/^r/, Appendix S.^Plan of typical court room New York Court House. 38 MARKS AND WOODWELI, : DAYLIGHT IN BUILDINGS 7377^/^'^:^^^^^^^?'^^5?'?^75''^''^^;^%75^^ ■^.yy^i Appendix 9.— Section through fourth and fifth floors at colonnade New York Court House. MARKS AND WOODWELI, : DAYWGHT IN BUILDINGS 39 APPENDIX lo. COURT ROOM DATA. I. Proposed new Court House, typical court room, as shown on plans. II. Proposed new Court House, typical court room, assuming lighting from outside windows only and floor area limited to main room (exclusive of wing for visitors' platform). III. Old New York County Court House. IV. Test Room. '^°"''' Floor Glass Ratio floor "■f"™ ^ ,. area area area to Vi.sual 'yP« Location (sq.ft.) (sq.ft.) glass area anglei 1 5th floor 2,059 661 3.12 to I 21° 45' 4th floor 2,059 572 3.6 to I 21° 45' II 5th floor 1,639 428 3.82 to I 27° c/ 4th floor 1,639 339 4,83 to I 30° o' III 2nd floor — Special Term, Part 3, 2,060 451.5 4.56101 15° 30' Special Term, Part 4, (Room 2i8) 1,279 120 10.67 to I 24° 30' Trial Term, Part 9, (Room 308) 1,768 148.5 II. 9 to I i5o45' IV Test Room 232 45.5 5.1 to i 21° 45' 1 Taken 3 feet (0.9 m.) above the floor at wall opposite windows. BIBLIOGRAPHY. H. B. Molesworth, Obstruction to Light. Wm. Atkinson, Orientation of Buildings, or Planning for Sunlight. Handbuch der Architektur, Bd. 4, Chap. i. K. Mohrmann, Ueber die Tagesbeleuchtung innerer Raume, Berlin, 1885. L. Weber, Intensitatsmessungen des diffusen Tageslichtes, Annalen d. Physik u. Chemie, Bd. 26 (1885), p. 374. La fenetre etudiee comme source de lumiere dans la maison, Trelat. Revue d'hyg., 1886, p. 647. C. L. Norton, Diffusion of Light through Windows, Report Insurance Engineering Experiment Station, Boston, September, 1902. Max Gruber, Daylight Illumination of School Rooms, Gesundheits Inge- nieur, Nov. 18, 1904; abstract in Ilium. Eng., London, vol. VII, Jan., ^ 1914, p. 30. ''O. H. Basquin, Daylight Illumination, Ilium. Eng., New York, vol. I, 1906-7, pp. 724, 823, 930, 1016, and vol. II, 1907-8, pp. 13, 123, 192, 284, 446, 518. S. Ruzicka, Die Relative Photometrie, Archiv. f. Hygiene, 1907. L. W. Marsh, Daylight Illumination, Trans. I. E. S., vol. Ill, 1908, p. 224. B. Monasch, Window Panes and Daylight Illumination, Ilium. Eng., London, vol. I, 1908, p. 368. S. Ruzicka, The Provision of Adequate Daylight Illumination in School Rooms, Ilium. Eng., London, vol. I, 1908, p. 539. 40 MARKS AND WOODWEjI,!, : DAYLIGHT IN BUILDINGS P. J. Waldram, The Measurement of the Relation between Daylight Illumination of Rooms and Sky Brightness, Ilium. Eng., London, vol. I, 1908, p. 811. L. J. Lewinson, Intensity of Natural, illumination. Throughout the Day, Trans. I. E. S., vol. Ill, 1908, p.' 482; ' - •■ • E. L. Nichols, Daylight and Artificial Light, Trans. I. E. S., vol. Ill, 1908, p. 301. P. J. Waldram, The Photometric Measurement of Illumination for Archi- tectural Purposes, Ilium. Eng., London, vol. I, 1908, p. 741. Thorner Daylight Illumination Tester, Ilium. Eng., London, vol. I, 1908, p. 505. • F. Pleier, Measurement of Daylight in Schools, Zeitschr. f. des Oesterr. Ingenieur und Architekten-Vereins, No. 2, 1908, p. 531. F. Pleier, Wall Reflexes, Zeitschrift fiir Schulgesundheitspflege, vol. XXII, 1909, p. 227; abstract in Ilium. Eng., London, vol. VII, Jan., 1914, p. 31. P. J. Waldram, A Standard of Daylight Illumination of Interiors, Ilium. Eng., London, vol. II, 1909, p. 469. P. J. Waldram, The Need for the Measurement of Illumination : Daylight and Artificial, Ilium. Eng., London, vol. Ill, 1910, p. 89. L. B. Marks, Principles and Design of Interior Illumination: Daylight, Lectures on Illuminating Engineering, Johns Hopkins Univ., 1910, vol. II, p. 649. James Kerr, School Lighting — Natural, Ilium. Eng., London, vol. IV, 191 1, P- IS4- L. Weber (Kiel), Some Remarks on the Measurement of Diffuse Daylight in Class Rooms, Ilium. Eng., London, vol. IV, ipn, p. 243. L. Weber, The Study of Daylight Illumination in Kiel, Ilium. Eng., London, vol. IV, 191 1, p. 669. Annual Report of Inspector of Factories and Workshops, Great Britain, 191 1, p. 239. Report of Heights of Buildings Commission, New York City, Dec. 23, 1913. A Short History of Investigations on the Natural Lighting of Schools, Ilium. Eng., London, vol. VII, Jan., 1914, p. 27. Replies to Queries on Daylight Illumination, Ilium. Eng., London, vol. VII, Jan., 1914, p. 34. P. J. Waldram, Some Problems in Daylight Illumination with Special Reference to School Planning, Ilium. Eng., London, vol. VII, Jan., 1914, p. IS, and Feb., 1914, p. 75. J. R. Cravath, Brightness (Daylight), Trans. I. E. S., vol. IX, 1914, p. 403. Daylight Illumination of Schools : Report of Joint Committee, Ilium. Eng., London, vol. VII, July, 1914, p. 359. CORNELL UNIVERSITY LIBRARY 924 051 402 950 GAYLORD BROS. INC. SYRACUSE, N.Y, 13221 LOS ANOELES, CA, 90060 MADE IN U.S,A.