the trading companies did not have an abundance

Kanegafuchi Boseki. At that time, unlike later on, the trading companies did not have an abundance of technical knowledge or knowledge of spinning machines and so, as will be explained in the next

section, it was the spinning mill's chief engineer or engineering manager that had the overriding say on technical matters. The chief

engineers went to England themselves in 1887 to choose and purchase spinning machinery (e.g., Osaka Boseki's Yamanobe Takeo,

Mie Boseki's Saito Tsunezo, Hirano Boseki's Kikuchi Kyozo and

Kanegafuchi's Taniguchi Naosada).22 Even if the result was that the

spinning machines were purchased predominantly from one manufacturer through the auspices of a particular trading company, the

initiative for decisions in choosing technology or machinery came

from the textile companies themselves and not from the trading companies. It was still some time before the trading companies would

hire university engineering graduates, act as a medium for detailed

technical knowledge or endeavour to provide good consultants.

A note about Platt Brothers itself is timely. By around 1887, Platt

Brothers had already managed to develop a new Rabbeth-type ring

spindle and it is thought that they had gained a firm foothold as a

ring frame manufacturer.23 It should be borne in mind, however, that

the company belonged to the group of latecomers in ring frame

manufacture. Although from early on Platt Brothers had a good

reputation for the development and manufacture of self-acting

mules, it was companies such as Samuel & Brooks and Howard &

Bullough Co. that became involved seriously in the manufacture of

ring frames. It was the superb engineers of Platt Brothers who were

responsible for catching up quickly in the manufacture of high quality ring frames. It does not follow, however, that a close relationship

with Platt Brothers would have stimulated the introduction of the

ring frame, had Platt Brothers been early starters in ring frame

manufacturing.

Next there is the argument that the switch to ring frames was

spurred on by the growth in imports of raw cotton. Although at first

glance this argument also has some plausibility, from the point of

view of technology choice, it seems that there lies some confusion

between the decision itself and adaptation that resulted from that

choice. As is shown in Table 4.1, the average count had reached 18

by 1893, the year by which it is thought that ring frame production

had become thoroughly established. Certainly by that date it had

become possible to produce finer yarns than had been spun when the 

mule was dominant. On the other hand, there were also large amounts

of raw cotton being imported from India at around that time, along

with the settlement of a special discount loading contract for Indian

raw cotton with Nippon Yiisen in 1893 and with a rising spun count.

It is necessary, however, to look back to 1889, the year when the

main decisions were taken, to link the problem of raw cotton supplies

and the switch to ring frames. It then becomes clear that it was by the

end of that year that small amounts of raw cotton started to be

imported from India (7 per cent of total raw cotton consumption)

and that the main sources of raw cotton were China (68 per cent)

and Japan itself (25 per cent)?4 As is well known, the quality of

Chinese cotton was on a par with Japanese cotton and it was therefore mainly suitable for yarns up to 17 counts. In other words,

production at that time was based on the premise that the raw cotton

would come from Japan, or from China when domestic supplies were

not adequate.

In 1890, Mie B6seki, Osaka B6seki, Kanegafuchi and others

attempted the first trial production of 20s yarn using raw cotton

imported from India. It does seem, however, that there was virtually

nothing known about Indian raw cotton by July 1889, when members

of the Ministry of Agriculture and Commerce and the All-Japan

Cotton Spinners' Association (Boren) left for India on a Indian cotton study toUr. 25 So it is rather difficult to argue that the ring frame

was adopted because it would be able to produce 20s yarn using

Indian raw cotton since the various pieces of evidence from around

1889 are not supportive of this view. To sum up, although a little

later on 20s yarn from Indian raw cotton came to dominate production, this development was a response to the adoption of ring frames;

it is rather difficult to consider it the cause of the switch to ring

frames. For example, as was still observed in 1893, the difference in

yarn counts that the mule mill and ring frame mill produced was not

really caused by the difference in raw cotton that was used: rather,

it emanated from the difference in the production systems themselves.26 In fact, the hypothesis that the switch to ring frames came

about because of the variety of raw cotton has an important implication for the decision to use mules initially.27 It seems to imply that

the mule was chosen to produce very coarse yarn and waste yarn

from Japanese raw cotton. This interpretation fits with the contents

of a record of the Country Fair lecture series on cotton yarn that was

given at the 1885 Gohin Country Fair;28 and it also accords with the

discussion in previous sections of this chapter. In particular the mule, 

which consumed little energy per spindle and was suited to smallscale production, would have been more appropriate than the ring

frame (or the throstle) if water power were the power source. Thus

the mule was a viable contender when the choice of technology was

being made, given that it could still be used for the production of

coarse and medium yarns made possible by improvements in the

quality of raw cotton.