Vol. 2 CONTENTS

5. the process of drawing
5-1. Steps of the process
Step 1. Draw two center lines
Step 2. Determine the framing
Step 3. Confirm the framing much more exactly

Step 4. What is the Horizontal line? and What is the Vertical line?
Step 4-1. about Vertical line
Step 4-2. for the Flog Perspective
Step 4-3. Why do I mention about the Horizontal line and Vertical line?
Step 4-4. the additional explanation about Horizon and a vertical line

Step 5. Confirm your eye position
Step 6. Determine the whole lay-out
Step 6-1-01. You need to understand the desposition of the object
Step 6-2. Measuring each object, not geometrical object,tree, mountain
Step 6-3. Checking out each relative positions and proceeding measuring

Step 6-1-01. You need to understand the desposition of the object for
Step 6-1-02. Measure the targets on each layer
Step 6-1-03. another examples about measuring
Step 6-1-04. another examples for measuring
Step 6-1-05. about our limitation of accuracy

Step 6-2-01. measuring about a tree on layer 1=near distance
Step 6-3-01. It is quite important that we are always checking out
Step 7. Observe the objects with binoculars and determine the figure
Step 7-1. Finish the layer 3=far distance part
Step 7-1-01. How to draw layer 3=far distance, smaller details?

Step 8. Finish the layer 2=middle distance
Step 8-1. How to draw layer 2?

5

he process of drawing

The process of drawing will be different for each case. And the process is just a summary and common procedure, so you can't finish your drawing only to trace the process, you need to examine the structure of your view and understand the relative positions and nature of what you are going to measure.

But you have to have above all sympathy with the trifle things and the ordinary scenes of daily life and the nature itseleves. Because to draw examiningly each object one by one to fill each up small part, needs paintience for long time work, and seems to be often very monotonous process without the ability to enjoy to find out even a trifle thing.

You can say easily or for just by a mouth without the sympathy like a saying "the world that I see is beautiful!", but you can't do it. You can't fill up even a bit, if you think a beauty is only in an idealism like a Raphael's painting or a scene of ancient Greek.

For my drawing scale was invented for dipicting exactly those details of daily life and the process that I am going to explane is just to do that.

What you can get by my drawing scale is only a exact lay-out of the view that hasn't details and contents, just lines and skeleton. Just your volition's that fill up the details and contents.
If you want to try by yourself this method, even though you haven't experienced about long time sketching on the field, then you had better to choose smaller sheet size, for instance, 260x130mm. The length "l" of the drawing scale can't be longer beyond you arm. (fig 3) If the length is same l=600mm as the instance (fig 16), then the effect of barrel shape becomes bigger according to the distance from the center to the peripheral, so the effect becomes smaller for a small sheet size.
I need 50 to 90 hours to finish for a sheet size, 260x520mm or 330x510mm.
The sheet size 130x260mm is 1/4 area of the size, 260x520mm. So roughly saying, I may be able to finish the smaller size for 13 to 23 hours or much more quickly.

If you had had enough experience, then you can do much more longer work that needs patience. But you will be lost your way on too long way beyond your expectation without a map as a experice, you won't be able to understand only what you are going to do at first time.
So you have to choose smaller size, if you want to try it only by your curiosity. Still you won't be able to do it , if you can't draw or paint even a simpler subject.

We can't advance but one step by one, especially for those that needs technique.
But I proceed to explain the process for the sheet size 260x520mm.

5-1

teps of the process




Step 1.

Draw two center lines.

This step isn't necessary, but I do it like a mechanical drawing for temporary use.
I draw two center lines exactly on center to the height and width of the sheet.
And I used to draw border lines surrounding the image size on the whole sheet of paper.

Landscape in Osaka: sheet size; 370x550mm, image size; 260x520mm
Landscape in Otsu: sheet size; 370x550mm, image size; 280x510mm/330x510mm

I have two reasons.
One is, it looks better for a drawing or watercolor to have blank margin.
The other is, I can scribe data, memo and signature or other scribles.

I don't like sign directly on the part of my image especially for landscape, because I draw precisely details, if I signed directly on the image, then my signature had hidden or damaged my details. My drawing or watercolor is itself my signature, so truely I think I don't need to sign, but there are many people who trust on much more a mere name by written letters than a work itself.
But like a still-life fish itself having much more blank margin with calligraphic letters is very effective and impressive by the combination the image and letters.

Step 2.

Determine the framing.
By my method, you can't contain the objects as you will. On this point, the method is the same as a camera by the view finder.
So you need to confirm roughly your framing of picture, which objects can be contained in your framing. As my drawing scale is matched the "l" to my reach (fig 3), so I can confirm roughly the framing by a scale 600mm, stretching my arm and comparing the sight with the scale of span W=520mm for the width and 260mm for height.

What I can draw on my sheet 260x520mm is roughly in the scale between zero and 520mm for widh, zero and 260mm for height. (fig 18)

Step 3

onfirm the framing much more exactly

Then you need to confirm the framing much more exactly. Because the particular objects on both side ends are often to give the important role or impression to the composition, so you need to cofirm exactly whether the objects will be contained or not.

And you need to remember, that the line a'b' is smaller than the segment a'b' (fig 4), and the line is the value by the above mentioned roughly measuring by hand, and the segment is the value by the measring by my drawing scale. So I won't be able to contain the important object on an end or both ends, unless I hadn't confirm exactly the framing.

From the end of right hand or the left or even the center, I measure by a dividers the distance between two objects (target A and B, fig 19) recognized clearly as a mark in the middle distance for depth, and each measured value (the distance ab) is better less than 50 to 100mm.
I can measure the value 160mm as the maximum value.


And this is very important for measuring, that you must never measure the distance between the objects on too much different distance (D1 and D2) for depth(target B and C, fig 20), becasue those may give often different value by the error of your eye position "E", and I can say it for the objects on near as well.

You can easily make it certain that fact, by changing your eye position to 'a' side, if the distance or the relative positon between the objects will be changed or looked different, then you must look out and measure another unchangeable or neglectable value between objects.

Of course, you can get the same value by measuring "bc" and also "dc", because point "b" and "d" are on the same line "ab", unless your eye position "a" have changed. (fig 20)

You know by the former examination(4-3-03-03, 4-3-03-04) that the value "bc" remains almost same for the error of eye position "E".
But the value "dc" will be changed from "dac" to "da'c" by the error "E". The value of differece "da'b" will become much bigger value by the ration D1/D2.
So you must never measure the target B and C, but A and B.
The former examination(4-3-03-05) can be applied for this case.
Still you may measure the distance "dc" when the ratio D1/D2 is nearly D1/D2=1 or target B, C are in far distant.
The point of the question is whether the angle "dac" and "da'c" are almost same by the error "E" or not.

lastly for Step 3, if you can't contain the objects as you will, then you need to walk around and find out another position with your tool until you are satisfied your compostion.

Step 4

hat is the Horizontal line? and What is the Vertical line?

There are only two straight lines as the imge of straight line in our view field, as I mentioned about (4-3-05) (fig 15). The other straight lines appear as approximately straight, or we aren't aware of its being curved.
Horizontal line:
I say it strictly, precisely, exactly, the Horizontal line will be looked out for the tangential direction on your eye position. And the earth is globe, then it is probably far in the universe.
But I am very generous man. I think I may say that if you can see Horizon, then it is approximately the Horizontal line that we seek.

If you can't find out Horizon, then what should you do?

You have to look out the substitution for Horizon by a binoculars.

See fig 21, you can see the roof and left side wall of building A. Then the roof of building A is below your eye position, i.e. Horizontal line.


You can't see the roof of building C, but the left side wall,and the top edge of the left side is converging to the center. Then you can understand that the roof of building C is above your eye position, Horizontal line.

So you can judge whether a line is Horizontal line or not, by the view of a roof and side wall.

If the roof of a building can't be seen, and the top edge of the side wall isn't converging along the depth, i.e. the top edge of the side wall and front wall are on the same level line, then its are on Horizon, i.e. Horizontal line.
Be caful of the judgement! The top edge of building B is also on Horizon, because we can judge it by comparing Horizon. If there isn't seen Horizon, you can't judge whether the edge is on Horizon or not. Because the side wall of building B isn't seen. You can see another building behind building B on center. Any side wall of the building also isn't seen about the building. So you have to examine the converging side by Perspective. If the top edge of the converging side is on level, then the edge is on Horizon on the center of your view fild.

When you can't see Horizon, then you have to find the substitution, the building at enough distance by your binoculars by above mentioned method. For the angle between Horizon and the building's roof at enough distance will be nearly unchangable by our eyes angle.

Step 4-1

bout Vertical line

I think I need to be explain about Vertical line. What I say as Vertical line is the Vertical line on the center of a view field. Indeed I say it again, the others are looked as curved on our image.
But, as far as we turn our head, or our eyes to the object that we are going to measure, the the object will be set on the center of our visual field. So all the objects that face to the direction of the nomal line on the cernter of our eye, will be seen as a straight vertical line.
And we can't measure the object without turning our head or eyes. So it is impossible to find the only virtical line as the case of Horizon.
So I declare that the Vertical line is on the center of picture. But this declaration isn't important, but just formal. Because I deal with all the vertical lines as straight vertical line, and it will be natural for a painting, I think.
And this process is only to confirm which building or object is on the center, or near it.
By the same way with Horizontal line, you can make it certain whether the line is on Vertical line or not.

For this case, you need to find out the converging side, the roof of a building. If one of the edge is on the same perpendicular line of the front wall, then it is Vertical line.
These explanations about Horizontal and Vertical line are, of course, assuming the Bird's eye Perspective, view from a hill or higher eye position. So I can judge those by the shape of roof.

Step 4-2. for the Flog Perspective

What should we do for the Flog perspective, low eye position on flat ground?

Then we can't see that important view, the shape of roof.
Honestly, I seldom have the Flog Perspective. Becasue only the near objects compose the whole view and hide completely behind its, don't make me feel the extention of space. And the composition is rather simple.

If there isn't the object like for Bird's eye Perspective,I have to find out the Virtical line for the mark on ground, or the row of roof tiles on slant roof, or the details of windows, or the scattered object on a Perpendicular line.

And when there is nothing, you can set the Vertical line on the center line of the picture, very simple!.

Step 4-3

hy do I mention about the Horizontal line and Vertical line?

I have to do so to avoid the misunderstanding about the view field.
The exactly straight Horizontal and Vertical line as our image on retina is only one for each, I think I may say again. And that is on the center of our eye's view field, so Horizon will be also seen as curved line for our eye's angle. You can see well that effect by the wide-angle lens of a camera for high or low angle shot.
So the image of only one straight Horizontal or Vertical line are chaninging according to two center line of view field being chaninged by the view angle, our head or our eyes angle.

So the straight Horizontal and straight Vertical line that I have explained are truely or simply two center line of the picture. But I think we can't accept a curved Horizon as Horizon except an unusual or a fantastic image. Then I declare that Horizon is a straight Horizontal line anywhere in the picture. This declaration is important especially for Horizon. For the Horizon is a important base line of a landscape for measuring. And I think the straigh Horizon must be the natural image for our psyhcological existing. And the other horizontal lines measured from the Horizon naturally gather up the barrel shape. Almost all Vertical lines can be deal with as a straight perpendicular line, unless the effect of perspective is needed, must be the natural image, I think.

But if you use a sheet in long side for height, then the Vertical line will be a base line on the contrary of the former explanation

On this Step 4, three points are important. Those are below.

1. Horizon is always drawn as a straight line, wherever Horizon will be arbitrarily placed on a sheet.

2. Vertical line is usually drawn as a straight line, unless the effect of perspective is needed.

3. We can make it certain whether any line is on the Horizontal or Vertical line or not by examining the edge of each converging side.

Step 4-4

he additional explanation about Horizon and a vertical line

We know that our landscape is the thin layer spread on the surface of the earth, no more than serveral handreds meters, except the mountains. I suppose that those daily understanding about our environment will affect our psychological recognitions.

However the earth is a globe, our usual sight will appear as exactly the flat land expansion as far as we see it on the earth, even the Horizon of the Ocean will be the same for our psychological recognition, though I haven't really see it once, for we had no true straight line independent from our eyesight as a standard, we had no certification for our sensation, but just we believed it, all the straight objects must be curved within our eyesight image.

So all the buildings stand perpendicularly on that flat land expansion might be expected to appear as the straight vertical line is psychologically the natural recognition as a picture.

Why do I repeat to mention that?

For , as I have mentioned, the image by the measuring by my drawing scale is shaped as the barrel image as spherical coordinates. ( fig 15 ) That means that all the horizontal lines and vertical lines except the two center line equally lean toward the two center lines.

However the picture is exact for the result of a measuring, we might never be able to willingly accept the clearly perceptible slant buildings and houses on the peripheral of a picture.

We can't control it for a photo, ( indeed a camera gives us a picture only to push the release, but once we pushed it, then the camera had finished its job indifferently about our interesting) but a picture, a drawing, for we have ourselves to draw it all.

So we might naturally ask much more the natural image for our recognition.
And it's much more easy to draw by treating every vertical line as a straigh vertical line.

Here, I give priority to the psychologically natural image as picture much more than the exactness of measuring.
Those declaration about Horizon and the vertical lines has to be understood on that meaning.

So there happen, especially on the peripheral, the disagreement between the measurement and the drawing, and that is caused by the different principle between the measuring by spherical coordinates and the actural process by the psychological recognition.

But I believe it is reasonable way.

Step 5

onfirm your eye position

We have confirmed the Horizontal and Vertical center line on Step 4. So we know now which object is on the Vertical center line.

Then we have to recognize where the position of our eye is.

Because when we measure the relative distance between objects on different distance along the depth, then we have to set our eye exactly on the center position, especially it is important for Horizontal center position. For our head position is always changing while we are drawing, unless we fix our head on something post. Though we know by the examination (4-3-2) that we need not to fix our head, when we measure each object or the distance between object on the same distance along the depth.

And we can't finish the landscape in a day, we have to work at least 10 days , 50 hours. So we have to confirm our position and exact eye position.
We can confirm our eye position easily by checking the relative position target A and target C, its alignment. (fig 20)

I have mentioned that the relative position is easily changed by the error of eye position. So we can check our eye position by that.

It is needless to say that the targets need not be always the building's edge. We can choice other objects or details, post, window, etc as well.
I use the bench rule with spirit levels, that I glued an acrylic spirit levels on an transparent acrylic rule. So I can confirm Horizontal level and Vertical line by the rule.(fig 22)

Step 6

etermine the whole lay-out.

Step 6-1. Measuring each structures in middle distance, then near, then far from center to peripheral.

Step 6-2. Measuring each object, not geometrical object, tree, mountain, river, etc by its characteristic points.

Of course, these lay-out are drawn by pencil for just out line, temporary line.

And the structures in far distance are usually very small in picture plain, about less than 10mm.

So I don't draw it precisely and measure except the important object, and leave them as blank. You have to examine and draw them with binoculars by eye measurement. You will confirm the same thing again, when you draw it with pen and for coloring, even if you drew its precisely with pencil, unless you left the pencil line as a line for finish.

Because you will probably forget what you have drawn, for its are too small and meaning less drawing as independant objects except the role to fill up the blank by its atomospher, just atomospher supporting the whole perspective effect.
And you need to understand what it is that you are going to draw, especially when you are to draw the determined line, unable to correct again. So I usually draw those very small details directly by pen as decisive conture line and coloring.

Step 6-3. Checking out each relative positions and proceeding measuring

As the lay-out proceeding to the finish, one object is to have many relative distance between the other objects, so you have to check its whether the relative position is correct or not.

What should you do, if the lay-out wasn't correct?
You had better to correct it, if the measuring has been just starting, or if it is the fatal error for you.
But you need not correct it and should compromise, if the measuring is almost finishing thruough the much troublesome working. No one will be aware of your mistake, if you don't say it.
No one can be, even if you said it.

This is a picture.

But if you don't like it, then you can correct it as well.

tep 6-1-01

ou need to understand the desposition of the object for a separated each layer as the row of the objects along the depth. (fig 23, 24, 25, 26, 27)




Each layer has depth as a zone, though I show each layer as a plain for just explaining on fig 23.

layer 1=near distance:

The value of measurement or the image or the relative position of the objcets on this layer 1 are easily chaged by eye position. Namely, Those are comparatively too near for the error of eye position to neglect it, I may say it less than 50m from eye position, as I have examined on 4-3-03.

Those have rich details and will give the most impressive part to the picture. ( fig 24, 25 )

layer 2=middle distance:

The valuse or the image or the relative position of the object on this layer 2 are hardly changed by the eye positon. Namely, those are enough far to neglect the infleuence by the error of eye postion,
I may say it much more than 50m from eye position.
Those objects have also enough details and concrete images deserved to depict, and may give the impression of extent as space. ( fig 24, 26 )

layer 3=far distance:

The objects on this layer 3 aren't changed by the error of eye position.
Those haven't any details but the contour, and have only the role supporting the perspective effect, the impression of the extention as space. The height or width of the objects are less than 10mm, usually 5mm or so. ( fig 24, 27 )


Of course, the image of the objects are continuously decreasing their size accoding to the distance, but I think you will approve my simplification about a landscape, and think it as the natural process, if you had experienced to try to express by yourself it.

We can easily understand that the view of objects are influenced by the error of eye position according to the distance between the eye position and the objects, by supposing the view from the moving train, that near objects are quickly passing by, but the middles are slowly passing, and the fars are hardly moving.

I think we are probably often faced with the same phenomena that we estimate too much only by the reason as those standing near. The near objects that appear big and occupying the large part of our sight are sometimes occupying only the small part as the truth.

Step 6-1-02

easure the targets on each layer




See fig 24, this is the view of fig 23 from the eye position.

Target A, B and C are belong to the layer 1.

Target D, E and F are belong to the layer 2.

Target H and targets are belong to the layer 3.


At first,
confirm (central) Vertical line and target B and H are on the Vertical line. And you can confirm whethre your eye position is correct or not by the alignment B and H on the Vertical line. (Step 5.)

Secondly,
measure "1"=distance between not target B and D but target H and D, because the distance B and D is easily changed by the error of eye position, I say again, then "2"

Thirdly,
measure "3" from the basic Vertical line=target H,then "4", then "5" and "6".

Fourthly,
measure the vertical value "7" from the Horizon as the basic line, then "8". You may measure each vertical value for each target.

Fifthly,
measure target A, B and C on the layer 1, "9" from target H, then "10","11","12","13"
When you finished to measure all the targents on the layer, then you confirm your eye position by Step 5 and measure the relative distance between a target on the layer and a target on another layer. And you proceed to measure the targets on new layer. Then you do again the same Step.

Attention!
I may suppose that you have been aware of the difference about measuring between the layer 1 and the layer 2.

For the layer 2, the measurings are based on the Vertical line,
target D from the Vertical line, target E from the Vertical line, target F from the vertical line

For the layer 1, the maeasurings aren't based on the Vertical line, but the distance between each target.
target A from target B, target C from target B

Why?

For example,

each horizontal value is

5=1+2+bc+4+de

The equation is correct, unless the each value contains the error.
But the measurement actually contains the error, so the value of each side is supposed as below.

The left-hand side= 5+error
The right-hand side=(1+error)+(2+error)+(bc+error)+(4+error)+(de+error)


Namely the right-hand side contains 5 error, though the left has one error.

If the absolute value of each error is less than 1mm, then the each side value is expected as below.

the true "5" - 5mm < the value of right-hand side < the true "5" + 5mm
the true "5" - 1mm < the value of left-hand side < the true "5" + 1mm

Namely, the errors are accumulated by the times of measurement.

But you can't always get the good result to measure everything by the basic line. Because if the each value of measuring become much more than 150mm, then the measuring itself becomes difficult, and probably cause the much more error as I mentioned on Step 3, and the problem between the segment of circule and the straight line arises, as I mentioned on 4-3-03-01.
We need to remember the nature of measuring based on the sepherical coordinates.

And as I mentioned on Step 3, we have confirmed the framing by measuring characteristic building or landmark or point, the value less than 50 to 100mm, we had better to measure another objects to check whether the measurings contradict those check-points or not.

See the fig 25, 26 and 27

The width of targets on layer 1 are more or less about 50mm. Usually these values are comparatively big, so we may measure those from next to each other.

The width of targets on layer 2 are less than about 20mm. We have to put neatly these small houses and buildings between the detemined distance, the above mentioned check-points, for example, the target D and E between the Vertical line and point "e" (fig 28)

We have to be able to determine which method is reasonable for the case. And we are learning by the practice. That is the experience. That is the basic rule. We need to apply flexibly the rule.

So the Horizontal line and Vertical line may be also replaced by another easy recongnizable or measurable line as a temporary Horizontal and Vertical line.

Step 6-1-03

nother examples about measuring







See target A

This is an example that each side of an object converges to a point on target B. (fig 29.)

First,
Confirm Horizontal line and Vertical line by a level.
You have to check whether the horizontal relative positions don't change by the vertical eye position or not. If its changed, then you have to straighten your back and measure.

You can make certain those below.
Horizontal line crosses with the vertical line "df" at the point "b", and you can memorize the point by the window's upper line.
Vertical line crosses with the window's lower line at the center of window. You can confirm your eye position by this relative position between window's center and a post .

So you can measure "1", "2" and "3", and the vertical line "ce" can be used as a temporary Vertical line.
Then you do the vertical distance "ac", "bd", "bg", and "ce", "df", "gh".
You can get the vanishing point by extending the lines of target A, "cd", "ef". And if you had drawn target B, you could confirm whether the vanishing point on target B is correct or not by comparing with the view with a bench rule.
By the same way, you can conform the extented line "dg" is on the corner of target F.
You can confirm whether your measuring is correct or not, to examine many relations with another target.
As the skelton getting nearer to the finish, you can use much more relations.

Step 6-1-03-01. A way to detemine the arrangement of the windows.

As all the elements, windows etc. on the plain "cdef" are converged and lined up for the vanishing point, then you can determine the position of each element and draw by one-point perspective.

For example,
if you want to divide the line "cd" for 5 equal portion along the depth, then you can get the 5 division lines by following process below. ( fig 29 )

Divide the line "df" for 5 equal portion, and draw the conveging line from each division point to the vanishing point, and draw a diagonal line "de".
The 5 division lines that you want are the vertical lines on each intersection point of the converging lines and the diagonal line.

Then only what you do is to dispose windows on each divided space. You did draw the window.

Step 6-1-03-02. Another way to determine the arrangement of the windows.

This is simple, you can do it to measure each horizontal and vertical distance, "4", "5", "6", "7". ( fig 29 )
Though I think it needless to say, don't measure the vertical distances from the Horizontal line, but from the line "dg" or "fh", for these windows are belong to the plain "dgfh", then each side can be the basic line.

Step 6-1-04

nother examples for measuring


See target l ( fig 29 )

Target l and a post on the left are belong to layer 1, near distance.
So the relative distances between target l, post and the other supposed to be easily changed by your horizontal and vertical eye position. Then you have to check whether the relative distances are changed or not.

The horizontal eye position can be checked by the position between a post and the window of target A.
The vertical eye position can't be changed beyond your body, namely you may just straighten your back, when you have to check your vertical eye position.

You can measure the distances from the post about their horizontal positions of points "i" and "j".

You can measure "8" from the line "ef" and "9" from the Horizontal line about their vertical positions.

You can also determine the position of point "k", if you had the determined two points. The intersection point by the circles, "R1" and "R2" is the position of point "k".

Step 6-1-04-01. Here I explane how to draw the roof tile of target l.

Step 6-1-04-01-1. If the size of tile isn't decreasing or hardly decreasing along the depth, then you can divide for equal portion.


Step 6-1-04-01-2. If the size is decreasing, then you can choice three ways below.

The first way
is that you measure every line.

The second way
is that you measure every 4th line, 4th, 8th, 12th, etc., then you divide each value by four for equal portion.

The third way
is that you use the intersection points by the diagonal line, mentioned about target A.
If you can get the vanishing point by two side of the roof, then you can deal with it below.

If the roof has 20 lines, then you have to draw 20 converging line and divide one side for 20, but it is difficult and not practical. Then you can use above mentiond second way, draw 4 converging lines and seek 4 intersection points for every 4th line.

If the each size of tile is much more than 5mm, then the first or third way is useful.
If the each size of tile is less than 4mm and much more than 2mm, then the second is better and practical.

Step 6-1-05. about our limitation of accuracy

You can't measure exactly each value, even if you measure every line.
And you can't get exact intersection point, even if you use the diagonal line.
You have to do those in the field, niether in your room, nor on your desk.

Can you exactly plot these value,

2.1mm, 4.2mm, 6.3mm. 8.4mm, 10.5mm, etc.

by the rule with the minium scale=1mm?
No!, you can't. And your pencile line itself has the width, more or less 0.3mm

I used to plot those value as

2mm, 4.25mm=4+1/4, 6.25mm=6+1/4 or 6.33mm=6+1/3, 8.5mm=8+1/2, 10.5mm=10+1/2.

Those are also not exact, but enough practical for a picture. And I think you will be able to divide 1 mm for two, tree and four parts, even though it isn't exact, has only 0.5mm as a result accuracy.

And, if the each size of roof tile is less than 2mm, then the house itself is small, namely enough in far distance, needless to examine the perspective effect, so you can divide for equal portion.

These explanations about measuring are basic way, and useful as a practical method, and I think you can find out and choice and apply the better way for each case.

Step 6-2-01

easuring about a tree on layer 1=near distance

I omit the measuring about river or mountain, for those objects are rather simple, I think I need not explain precisely for it, but just to briefly mention to measure several points about their horizontal and vertical positions ,and you only plot its, then draw, so you can get the out-line. Then you proceed to depict the detail of contour line by looking and examing with binocular along that out-line

See fig 30

This is a cider. I traced its contour line, trunk and limbs by heavy yellow line, and the out-line of houses and buildings by light yellow line.

The characteristic points are plotted as red point. The points are grouped into three kinds by the way of measuring.








First group
the points below the Horizontal line are usually determined by the horizontal and vertical relative position with the objects surrounding the cider, houses and buildings as available basic lines.

Second group
the points between A, B, C, D and the Horizontal line are determined by the vertical relative position with houses and buildings, and the horizontal relative position with the center of the cider, the trunk. So you need to have confirmed which limbs are on the center line of the tree. We can usually use the trunk and the limbs for the characteristic points to determine the position of each details.

Third group
the points above the points A, B, C, D haven't the available objects near its, houses and building as the basic line, but are determined by the distance from the determined two points with a compass, as I mentioned on 6-1-4.

The characteristic points that I mean have to be easily clearly distinguishable from others at any time by yourself.

Though I said about only the points on the out line, you can choice many points for each branch,and you need to understand the tree as not only the outer shape but also 3D objects constructed by each parts, the trunk, limbs and branches with bunch of leaves.
Those recognition will help you to depict the details, shading and coloring.

When you have finished the measuring of all the points that you need, then you observe the details with binoculars between two points and depict the details of contour one by one.
You need to observe carefully about each two points
whether the tendancy of the contour line is concave or convex,
then how many inflection points it has?,
whether its interval is regular or not?
how's its size between the top and the bottom comparing with the distance between two points?

Much more questions you have about the contour, much more precisely depict the details you can.

The depicting about the details below the Horizontal line will be easier by comparing with the details of the objects surrounding the tree.
For example, you can use the roof tiles as a grid.

By the same way, you can measure the tree on middle distance. Then the measuring points will be fewer than the tree on near distance. And much more far the tree located, then you may just measre the height and the width. It's easy.

Here, you have finished the measuring about main objects.

Step 6-3-01

t is quite important that we are always checking out the available relative positions with others and proceeding the measuring, as I have mentioned on Step 6-3.


We need to check out these three points.

Point 1
the horizontal relative position with other objects
What objects are at the both side of the object?
And those are really the same as the sight?

Point 2
the vertical relative position with other objects
What objects are above the object? and below?
And those are really the same as the sight?

Point 3
What objects are on the extention line of the object?
This is quite useful to confirm or determine the slant of roof, converging line by Perspective.
I used to choice the distantest object on the picture plain to get rid of a plot error. (fig 30)
But you need to be careful to do that, for you need to considerate the effect of barrel shape by the spherical coordinate measuring, as I mentioned on 4-3-4, (fig 15).

Especially you need to be careful for the peripheral part of picture, because the curvature becomes bigger according to the distance from Horizon and Vertical center line. So I have explaine again and again about Horizontal and Vertical line.
Indeed I have confirmed the object on the extention with a straight bench rule by my sight, but the sight is sight, the picture is picture by the shperical coordinate measuring, naturally shapes the barreled grid for two straight center lines, Horizon and Vertical center line.
about Horizon and a Vertical line.

More or less you will be faced by the disagreement with the measuring, especially on peripheral part of picture, four corners. Then you need to remember the effect of barrel shape, and seek the practical solution.

See fig 31,
I could confirm with a bench rule that the house C and D were on the same straigh extention line, but I think that extention line may be curved on the picture plain by the effect of barrel shape, because the extention line is far below the Horizon. Namely I suppose the angle of the roof line C is getting probably smaller to horizontal line as it is getting much more far from the Horizontal and Vertical center lines to the peripheral. The space of A and B are actually much bigger than the value of measuring by the abovementioned reason.

I have mentioned about the cause of that disagreement on Step 4-4, the aditional explanation.

You may soon be aware of the fact by seeing the extention lines on fig 31, that you can never construct the whole view of a landscape by the artificial drawing method, like One or Two-points Perspective.

Because the arragement of real houses, buildings and structures aren't always built or deposed at right angle each others, on the contrary a regular desposition seems rather exception, so you can never apply such a simple Perspective for a real view.

Still you can apply the One-point Perspective partially for the groupe of houses or buildings, and that may help you understand clearly or logically the desposition by the simplification.

See fig 32
The house "A" belongs to the vanishing point "a".
The house "B", "C", apartment house "D", buildings "E" and "F" belong to the vanishing point "b".

Here you have finished almost the process of measuring, I think you will have spent 40% of total amount hours of the work only for the measuring.

For example, I think I needed probably 40 hours/ 4 weeks, 10 hours/2 day/week, for the measuring about the Landscape in Otsu No. 20 (fig 31), the total amount hours of work 92 hours.

I gess you think it too much only for measuring.
But I think these troublesome work is not useless. Because you will have observed the changing of lighting, weather conditions and the variety of appearance of the view until the finish of measuring.
You will be gradually constructing your image of your sight. Those observations will give you enough imformations about the view, make your image rich with the details, and heip your decision about the next process, shadowing or coloring.

I think this steady working is quite different from a photo, a picture is the image of an average impression during a particular period of time, but a photo is the image of a momentary impression that gives us sometimes very diffierent and unknown impression from our own image about familiar sight.

If you have luckly finished the measuring without any contradiction, that will make you feel simply the great enjoyment about your exactness.

But what you can get by the measuring is bare an skeleton drawing with straight lines by rule and a bit of free lines by hand, that, I suppose, doesn't look like a picture for you but a mechanical drawing, or an almost blank sheet with few line.

Actually, I think you can't recognize anything but a blank sheet, if you see it from more than 1m distance. And it is really a horrible sight that a man with just a trunks is sitting still and silently in front of an blank sheet of paper for 5 hours, almost motionless and is staring at houses with binoculars from a hill for much more than a month, I think by myself it really horror, still, whatever people felt and thought, indeed I was by myself greatly enjoying it.

I can suppose that everybody can get almost similar result, about same framing, same view untill here, if he had have enough experience about my measuring method. But you can choice many different expressions from here even for same realistic landscape painting.

Everybody can say it easily only by his or her mouth or by his or her camera that a realistic landscape is only to reproduce the view. Who can reproduce it? Such word has nothing for contents.

I will explain each expression by each example, along each my work. So I explain from here only the process briefly.

Step 7

bserve the objects with binoculars and determine the figure of the objects and depict the details by pen.
You need to finish the pen drawing first about the layer 3=far distance, then the layer 2=middle distance, at last the layer 1=near distance.

Those reasons are below.

First reason
once you had drawn something by watercolor, then you can never erase it again, and you can't lighten your watercolor, but only darken it.
Namely, most charming and vivid parts as the subject of the picture should be finished at lastest, because its are usually the most bright and high contrast parts. The other parts should be colored according to the order of the whole.

Generally, we sense the brighter and higher contrast object being nearer than the dark and dull one.

Second reason
the working to dipict the layer 3=far distance is often monotonous and troublesome, needs patience, I think, can't be done sufficiently after the enjoyable working about the layer 1=near distance.

If you did your best about that monotonous parts, the layer 3, then you can do naturally much more better finish about the layer 2 than the layer 3. I think you can't do that by the reverse order.

I haven't anything to explain as a general rule or process about expression, for I am now seeking the best solution too, and if I thought I had found that best solution, and if I was applying that same process like a sterotype, I think, my watercolors had been nothing but a mechanical process, like a camera fix an image on film.
A colloapse then begins and ceases to be as an art, perhaps. An art is curious.
Am I doing that?

Step 7-1. Finish the layer 3=far distance part

If you choice the fine day as the lighting condition and you want to finish your watercolor by full coloring, you need to examine carefully and remember the shadowing of objects, and its color.

And I think it better that you finish the details part by part completely, if you can. But I think I can say it honestly that you may need quite enough experience to do that.
You may do the coloring after when you have finished the drawing, if you have enough time and patience to do that. For you need very long time waiting the same fine day for coloring and to maintain your concentration to your watercolor until the finish.

If you want to finish your watercolor by monochrome shadowing, you may finish the drawing at first, then the shadowing. For this time, you need to be careful only for the shadowings of objects. It is much more easier than the full coloring.
So if you want a steady process step by step, then you had better to practice the monochrome, then the full coloring.

Anyway, I think I can say that the solution that now I use for the expression about the far distant view is reasonable, so I explain here it as one example.

See fig 33, 34, 35, 36
These are the details for the example expressions of layer 3=far distance

These are watercolors, the material, watercolor itself limits the expression.
And the sheet size of these watercolor is small and it also limits the exprssion.
And you have to do these work in the field, open-air limits your working as well.

You need to consider these conditions.



I depicted the detail with brush by eye observation and measurement. (fig 33) I didn't use here the contour line for dipicting details. So the expression is primitive but much more natural than later works, I think.
See the detail of the window of a house below the scale "13". I didn't use contour line to depict the sash bars but expressed it remaining as blank, papaer's white, and windowpanes are colored with grey. The parked white car in front of house below the scale "20" is also dipictied by same way, the figure is expressed by the color of surroundings and its window.
These expression without contour line needs much more working times than that you might suppose.
But I can't apply these expression without contour line for much more complicated landscape.
Because I have to express one side of a house with the different value of light or color that make you understand the boundary with the other sides or surrounding objects.
So the condition of lighting becomes essencially much more important to the expression.
Indeed I think I have said many times about that, you need to wait coming the same condition or you can't unify the impression of your picture, like one detail is high contrast, the others are dull, then each detail may give confusing impression each other.

I used usually Whatman paper until 1991 because of its whiteness. But Whatman papaer's surface is very weak, so you can't use eraser not to damage the sized surface of the paper.
So I didn't do rough drawing by pencile except a few slightest marking that needed not to be erased. I hadn't enough time for landscape painting as 1989, so I painted my watercolor directly by watercolor with a brush since 1990 to 1991.

See fig 34
This is the first work by my drawing scale. (fig 2) My drawing scale is very useful for such detail. You will be able to measure it perfectly. That details are depicted by contour line. The size of small windows are less than 1mm. You can't express those small details by brush, and even if you did it, you can't make someone understood what you wanted to express. The result will never compensate your troublesome labour with that effect.



I think a contour line is the simplest and most effective meaning to explain what it is for like those small watercolor.

Here I used to keep a pen point sharpen to depict very small details. I think you can understand that expression, that each smallest object itself is depicted by only contour line, has only their out-line and windows, has nothing flavour as a independant object for a picture, but is gathering up as the town probably to successfully make you feel the atomspher.
What I wanted to paint as I felt is also able to express by the object itself existence.
Here, what I did is only to depict the objects that I could do. I didn't find any separation between my sight and feeling, between A Subjective and An Objective.

For such distant object, we can't find anything but the contour line. If you keep the ratio of the size between each object, you can't depict anything but the out-line for those small objects and shadowing.

Jan Bruegel elder is known by his detailed depiction, but I think he mistook the ratio, especially his persons in his picture, he painted those for too much bigger to depict their details and by too much higher contrast and color. So that gives us strange and confusing impression of the space. I think he had to deal with those miniature persons by the expression like a Monets.
He can never be matched to his father only by that details.

Step 7-1-01

ow to draw layer 3=far distance, smaller details?




First,
you measure the position of the outstanding buildings for others.

Then you are depicting the smallest objects one by one, puting into neatly between the outstandings.

You need at first to recongnize the shape of each smaller object with binoculars, what it is and how it is?
If you didn't recognize the shape, then you can't depict it. You have to measure the object that the size is much more than 5mm, or it is better.

The vetical value is very clear and easy ,here I can use the bank line of Yodo River for the basic Horizontal line.
But I need patience to measure the horizontal value, still I can use each side of outstanding buildings for the basic Vertical line.

When you have measured the outside dimension of the smaller object, then you depict its details directly by pen and watercolor, the row of windows, observing with binoculars, by eye measurement. If you have determined the lighting condition, you have to examine where the shadows are casted on and you give the shadows.

For example

I think I probably choiced the lighting condition as 3 to 5 p.m. in fine day. I used to work from 10 or 11a.m. to 5 or 6 p.m. untill 1994. I was only measuring and drawing the conture lines until 3 p.m. coming the same lighting condition, or coloring the part not influenced by the lighting condition, after I had finished the measuring about the lay-out until the Step 7.
After 3 p.m., I was very busy, I have to measue, draw, depict and also do the shadowing always watching the shadows moving.
Because the shadows of evening light are quickly moving and passing on the structures and the earth. You have to be careful and hasten to depict. You can't understand the situation only looking the other painting.

See the bigger building below the scale "24". This is the building of Nippon Paint Co., LTD. The building casts its shadow on another building below the scale "18", that is the building of The Mainich Newspapers, or, the shadow may be the building itself's, for I remember both wings of the building bend at 45 degree to the center part, if I didn't misunderstand. (fig 35)



What I want to say here is that you need always to be careful of each shadow moving, especially before and after the best lighting condition. Those shadow makes the details much more vivid and gives the solid image to them.

You have only 2 hours to be able to use for the shadowing for a day. And much more smaller the object you are going to give the shadowing, much more time you need to confirm again after the drawing, which building that you have drawn is. So, for such smaller object, you need to proceed the measuring, drawing, depicting and shadowing at the same time not to waste much more time, before forgetting the memories.

So I said it is better to measure, draw, depict and finsh at the same time about layer 3 a part by a part on this Step 7, as I mentioned on Step 6-2.

And I think you have understood that you can express sufficiently only by the contour line and shadowing for the smaller details.

See fig 35.
I can say same thing about Landscape in Osaka No. 10, except I used the heavier line and much more definite expression, higher contrast than No. 6.
You can see the surface of Yodo River between both side of bank, that reflects the silhouette of buildings, and a boat is being driven.
The river reflects the silhouette, and that means there wasn't wind blowing. If wind blew harder, then the river was reppled so that the surface reflect regularly light like fragments of diamond. The impression of a river, sea are very changealbe by the condition of weather, wind and light, because its reflect the surroundings.

I don't know whether you know those things or not. But you have to determine a condition from several choices and assemble them for reasonable order, because you can't do it like a camera at a moment.

I can suppose that you judge easily those detail as the common result that anyone can depicted as only those were. Of course those had been there, but we aren't camera, we have to be reading and choicing one by one, for limited periods, by our taste, in reasonable order.

This is painting, As we can take personal arragement, permutation from only 8 characters, like a password, we can naturally choice many things, so many things from a view than that password.

Actually you don't paint the same picture as the others for a same view. That isn't mainly the result of your technique itself, but rather your reading and choice and volition about that view, that the fundamental elements determine the technique

I think I can say that a painting is similar to a reading, nobody say that he has read and understood the book just becaue he bought it, or he has been watching a TV just because he turned on the switch.

The state, reflecting the view on our retina is just to buy a book or to turn on the switch, we can go out left the book on a table or left the TV being turned on, it's just reflecting automatically without our consideration by our brain, or somebody say he did?

Namely, the image on our retina is almost same for evryone, can never mean that we can paint realistically the view as it on our retina. There are many choices and enjoyment how to express it by limited means and our ability. A man who says the realistic painting is only rerpoduce the view, says only about the state that the image being reflected on his retina, does never think how to take the image out from his retina on a paper.

So you have to always be prepared for shadowing and coloring which condition you should choice from the begining of measuring. You have to read and learn by the view, or you can't depict the details, I think.

I felt sometimes that a monochrome coloring is much more suitable for a landscape watercolor that has many those smaller details than a full coloring.
For I can read those attractive smaller details and also explain that eloquantly. Because a color itself has independantly a particular order for a Perspective. And that makes the Perspective effect confused.

For example, Red is coming nearer than Green by the impression at the same value of chroma.

And the nature of the watercolor material itself makes the coloring difficult.

For a higher chroma color darken the picture. So you can't get the same result as the oil, acrylic painting materials and gouashe, able to cover the basement completely by the material.

But watercolor is still attractive material for those characteristics, quick dry, high quality pigments, above all never damage the texture of paper.

I can paint much more realistic, if I use gouashe. Because it has strong ability to cover the basement, that doesn't darken the picture even by using high chroma color, namely easy to control full color painting.
But the ability to cover the basement by the material naturally cause to cover the texture of paper by the substance of gouashe, the harmful damage to the texture, like spreading and pasting mud on a sheet of clothe.

Watercolor paper itself has the high quality texture can never be comparable with the other basements. Since 1991 I use Aquarell Arches Satine, has very fine texture. This paper has quite strong gelatine sized suface and exquisite and excellent texture. Gouashe covers the texture. And this is main reason not to use Gouashe, indeed, I don't like it, it is too heavy.

So you need a solution to handling watercolor for the natural expression. You can quickly recongnize it, what I said, as you have done it.

For example,
if you spread manganese blue or cobalt blue over a sheet, deeply impressed by the fine blue sky, then you will find it has turned to be dark night sky, very dark, almost nightmare.
And you can never make it brighter, without scraping and getting rid of the whole surface of the paper covered by the pigment.

See fig 36
Surely, here are the problems that I have to conquer, still remains.



Anyway, I depicted the smaller houses only by the contour line and gave the color just to cover the contour line. And the line is heavier.
These far distant parts in fine day are generally losing their color to violet-grey accoding to their distance. A green and a brown building are only given their color.
The roof tile of smaller houses itselves are black. But if you paint them with black, then their contrast will become stronger, then their color give the impression their being much more nearer than their rational distance.

And I think I can express sufficiently those far distant parts' color only by the color contour line.

And if you painted and cover the whole those part by their color, those became too darken and the impression of the picture became too heavy.

You can see the whole image of No. 20 on fig 31.

This is one of my solution for watercolor, that I determined I make the color of each contour line represent the color of each samller object itself and keep the picture bright by remaining the blank space, the material white of paper.
So I used the heavier line to be able to much more impress their color.

Watercolor materials have the diffirent values of their ability to cover the basement. Those nature also make the handling difficult, I think.

And especially watercolor material contains water as the medium, that will be vaporized soon, then the bare surface of paper will appear on the part where water had bee there.
Namely, the conditions of reflecting light are different according to being wet or dry.
The brightness of watercolor painting is depend on the materical white of the basement, paper. Light is reflecting on the surface of paper through between the pigment particles of watecolor material. And light is also reflecting on the pigments, that make us the color.
So the impression of color is the synthesized two sensed colors, the basement and pigments, like, black dots painted regularly on a white sheet of paper makes us feel the grey sheet of paper. Everyone probably knows it. I can say it about color too, like a principle of TV.

Oil painting material, acrylics, gouashe, these materials reflect whole light on their surface or their layer of material, indeed oil painting is also depend on the brightness of basement, but its medium, linseed oil is never vapourized, but polymerized with oxygen, namely the condition is less changable whether it is dry or not than watercolor, and if it contains few linseed oil, like impressionist method, then it reflect light completely on its surface.

You may be only careful of the brightness for monochrome coloring. But you have to be careful of two value, brightness and chroma for full coloring. And watercolor material is difficult to deal with for such values.

So you need to devise the expression.
I think there are main two points for watercolor.

First point is to keep Brightness for high chroma.

Second point is to keep Readability, especially for those that contain many smaller details.

you can't get the satisfactory result for brightness only to paste high chroma watercolor material, as I have mentioned former, for it only darken the picture.
And I think that an usual watercolor painting method, Wash, or Wet on Wet is helpless for realistic expression and detail depiction, except for sky. And that are the typical sterotype and obstinate idear about watercolor, that people think easily watecolor means only Wash or Wet on Wet expression, that I hate, not for the method itself, but for their obstinacy.

I think the dipiction by color-contour line is very usefulf expression for that two purposes. Because I can keep the Brightness by color-contour line to remain the blank space, the whiteness of the paper, and the contour line is effective to the readability of the details.

See fig 36,
how to draw the details
Fundamentally, the Step is the same about 7-1-1, here first you have to measure the buildings that you can recongnize easier than the other smaller details.

There are 6 buildings surrounding the smaller buildings or houses, like a colony or an island.
At first, you measure and draw the boundery, then you draw the inner part one by one observing the objects with binoculars, you may measure the smaller details and you may do by eye measurment, both you can do.

You need patience especially to depict these smaller parts, for you can't quickly draw them but very slowly and in steady way, and you need to sometimes change the color to draw the contour line of each part, and those changes are very small part and slighter and modest for your labour.

See fig 36,
the part below the scale 17 to 19, just below the mountain
I didn't use even the countour line to depict those details to make you feel the distance.

I depicted it with brush and gave them the light and shadow by the grey Wash, and I used vertical Hatching to give the solid image and the shadow to the mountain

Step 8

inish the layer 2=middle distance

You can find out many enjoyments about objects on the layer 2, for these parts are enough big to depict their details, you can feel the tangible quality about them.
So, if you saw something, then you need to recognize clearly them, what it is, with binoculars, and you have to depict them.

The important thing is that you must never judge which object you depict or you don't, by your taste or the evaluation about their utility, but you have to depict all that you can see and recognize, everything.

Don't bring your taste and the evaluation about the object itself into the picture!

We have to choice the weather condition, we have to choice the moving objects, we have to choice the expression, these are influenced by our taste, those are naturally the matters of the picture. But the objects are object independantly there, by their order, we are going to reproduce it, so we must never confuse it by our taste, unless we can't do it.

Step 8-1. How to draw layer 2?

We need to apply some stylization for layer 3. But we need less stylization for layer 2.
Generally, smaller an object becomes, stronger stylization we need to depict it.
Becasue we can't reduce it to keep every ratio about its details like a copy machine, computer graphic, camera.

So I can say here, simply, "Depict as what you are seeing"
And you can see the shape of each parts about layer 2.
And you can get each figures by the above mentioned method of measuring by the drawing scale, if you can't draw it by eye measurement.
You can see clearly the shape of each objects for layer 2.
And you can see slightly also the tangible quaiity about that material.

So you have to depict not only the shape but also the tangible quality, the texture of the surface for the object on layer 2.



See fig 37
The object on layer 2 is enough big, so that the object depicted only by the contour line can't have reality without the tangible quaility.

You can see two cars faced on each other on the right hand lower coner.
I think I depicted the cars very well, you can feel the tangible quality, its smooth surface of the body, metalic glare, front window, headlights.
I remember the color, metalic silver, silver grey. I depict the bonnet with gradation, so the left hand corner reflect light, and each corner as the highlight part.

The roof tiles on the center slightly change its tone at the quarter from the left side. You can't finish the trees only by the contour without the tangible quaility, its leaves.

There are a man and woman, probably his wife and children, their family. The man lifts up his dog on the shoulder. And a little girl is riding her small bicycle. And a cat is sitting.
I really saw those persons and a cat. I have seen peoples coming here to enjoy their walking along the bank of Yodo River on Saturday and Sunday. I felt it lovely, so I put those objects into my watercolor.
I am constructing the details by adding those persons one by one. Of course, you have to remain those space blank where the persons should be painted on, until you have finish its.

These persons in this watercolor, will soon disappear from the framing, are your choice.
You have to build up you image of the scene by your observing as a much more characteristic, suitable view. You can see the logo "TSK" on the top of a apartment building. I remember the representative "TSK" was to be soon summoned in Japanese Diet for his company's bad debt.

See fig 38
Here you can see the fence made of concrete block. And the number of block is exact. And you can see the part of tiled wall at left hand. And I counted the number. And the disposition of dark color tiles is exact, though I might miscount it for one line, or so.
At the gate two littl girl are playing, one of girl keep a kitty by string. She was very bad, hung and swung the kitty, a poor kitty. This apartment house was replaced by an apartment building later.

See fig 40
I like the dike, that is coverd and piled up by stones. The stones are irregular shape each other. The characteristic feature is well expressed, I think. The dike is built up by the combination with smaller stone and rather big one, but that top is neatly leveled.
Those are countable, one by one. It seems really troublesome, and was really so.

You will be able to do it,

then,

why don't you do it?


See fig 38

This is the prefabricated apartment house. I think the panels were undulating, so I gave the patial wash to the front panels.



See fig 39

I like these details, very crowded houses, and the confusing, but there is a daily life. A train, Keihan railway's is running through between the houses, is coming out and showing its head just behind houses. There are a chair and ladder in front of tin roofed house behind a green roof, a man is crouching forward, doing something at his small garden. A dog is kept, laundries are hung at vacant place at the right hand lower corner. A man is coming pushing his bycile and behind his back a moter bike covered with blue sheet is seen.



fig 40. Landscape in Otsu No. 7(detail) Jul. 1998
See the dike covered by piled up stones, irregular shape. I might be proud of it, not because of my technique, but because my attitude being sincere to the object itself. That arragement is exact. These are countable. If you had counted up and drawn its, then those were gathered up and naturally showing their character. I don't need any style to express it but just to draw it as those are.