Though we have so far spoken of reflected light mainly as it influences cornice tones, it should not he supposed that it has no effect on other values, for it has, though as a rule the horizontal planes seem to catch more such light than do the vertical. A window soffit, for instance, is often quite brilliant as is the intrados of an arch, while even as large a surface as a porch ceiling is often visibly brightened.
Before we drop our discussion of Figure 33, attention is called to the variety of methods of suggesting roof shingles which it shows. Too much care cannot be given to such representations, for in a drawing of the average residence so large an area is taken up by the roof planes that unless they are well handled the effect of the whole drawing may be ruined. First of all the values of the different parts of the roof must be decided upon, as some portions can perhaps be left white or nearly so while others will appear quite dark. Next, the method of indicating the roof material must be chosen, and it is here that the sketches on Figure 33 may prove useful, or. if the roof is of some other material. Figure 34 offers some suggestions. Sketch "A" on the latter plate represents shingles. flat tiles or slate - Sketch "B" indicates a rough textured slate in graduated courses - Sketch "C" shows shingle thatch, "D" straw thatch. "E" suggests tile, while "F" again shows slate, though a similar indication would answer for shingle. It may be well to mention here that good pen renderings are sometimes of great assistance when drawing roof or wall surfaces as they offer much in the way of material indication which can be adapted to pencil rendering.
There are several faults frequently found in representations of roof surfaces, concerning which the student should be warned. First of all, if a drawing is small in scale one should seldom attempt to show every course of slate or shingles, for if this is done the value is almost sure to become either too complex or too dark. It is better to space the lines separating the courses somewhat further apart than they would be in the actual building. In larger drawings this criticism does not hold unless the roof pitch is very low or the roof planes greatly foreshortened, in which case a small number of lines may prove sufficient to suggest many courses. When graduated courses of slate are shown as at "B" Figure 34, decreasing in size from the bottom to the top, an unpleasant effect of curvature of the roof sometimes appears. Such an effect, if conspicuous, can usually be overcome by throwing a shadow bounded by approximately straight lines onto the roof, as from a tree or some neighboring building. In fact, the addition of any straight lines following the pitch of the roof will help to correct such distortion. Whatever material is used as a roof covering, avoid breaking the tone into too many conspicuous spots, for one of the most common defects of the drawings of beginners is the spottiness of surfaces which in actual buildings would be either "flat" throughout or gradually graded.
Just as roofs deserve careful attention, wall surfaces also need to be represented with the greatest care. Here again it is seldom advisable to try to show every brick course or each stone but the materials should be so indicated as to leave no doubt as to their nature. Figures 33, 34, 35 and 36 all give suggestions for the treatment of such surfaces, the larger drawings on Figure 35 being of sufficient size to show the detail very clearly.
There is no great difficulty in acquiring the skill to render a wall of brick or stone, or a roof of slate or shingle, but when it comes to successfully representing windows or glazed doors or any objects containing large areas of glass, our task proves less simple, for glazed surfaces are so complex and changeable in their appearance as to demand special care and skill in their indication. It is not hard, to be sure, to learn to draw a typical window or two, especially if shown at small scale, but if the scale is so large as to make any considerable amount of detail necessary it is no easy task for the beginner to do even this much well, while it is still more difficult to so render a number of adjacent windows as to give them the best effect in relation to one another and to the remainder of the building. If they are made too dark or too light they may, even though good in themselves, attract more than their proper share of attention, and if all are drawn in the same way the result will probably prove monotonous, while if, instead, too much variety is shown, the breadth of effect of the whole drawing is almost sure to be destroyed. Before attempting finished renderings of windows the student should, therefore, acquaint himself through observation and study with the appearance of glass under different circumstances and conditions, for it is only by so doing that he can represent it to the best advantage in any given problem. Walk along a street and study the windows that you see, - not only those near at hand but those in the distance as well. Compare those on the sunny side with those in the shade, and those in the upper stories with those in the lower. As you make these comparisons ask yourself such questions as the following: What is the difference in the appearance of glass in sunlight and in shade? Do windows in the upper stories have the same general effect as those in the lower? How do windows in the distance compare with those near at hand? Can you see the curtains or shades distinctly in all the windows? How much of the interiors of the rooms do you see as you pass? Is the glass always plainly visible? Is it hard to tell if panes have been broken from a sash? Is it easy to distinguish plate glass when you see it? If so, why? Do all the lights of glass in one window look the same? Does the glass usually seem lighter or darker than the sash itself? Do you see images reflected in the glass? If so, are they sufficiently definite to permit you to tell trees from buildings? Does your own image appear in the windows? Are images more distinct in glass in shade than in glass in the sunlight? Are reflections as clear on a rainy day as they are when the sun is shining?