On Jan. 25, 1915, the inventor of the telephone, Alexander Graham Bell, inaugurated U.S. transcontinental telephone service.。
胡適之先生一生見證在橫越太平洋交通上的進步。
在電話建設上，中國相當落後。 譬如說 四川直至1990年， 每百人的電話約0.8具。
20世紀中國在電話建設上相當落後；Invention Factory：The Bell Telephone Laboratories, 1931
A Reporter at Large
New Yorker
NOVEMBER 28, 1931 ISSUE
Invention Factory
BY MALCOLM ROSS
TABLE OF CONTENTS
CREDITPHOTOGRAPH BY ALAMY / EVERETT
The Bell Telephone Laboratories, which is the research department of the American Telephone & Telegraph Company, is a good specimen of the modern invention factory, and handy to anyone who enjoys scientific peepshows. The ten-story main laboratory has fifteen acres of floor and takes up the block in West Street between Bethune and Bank Streets. Depression or not, the Telephone Company still finds it pays to spend nineteen million dollars a year there on research.
Except for the absence of elm trees and varsity letters, the Laboratories are something of a university, with eighteen hundred college graduates, seventy doctors of philosophy, and a crew of assistants, a total of five thousand people. Most of these are in the West Street building; but outposts are scattered from Colorado to Portnafrankagh, Ireland.
That seems an impressive outlay for a nickel in the Telephone Company’s slot. It appears, however, that not all of the nineteen million dollars goes immediately into improving the service. When a research worker gets on the side track of electronic physics, they tell him to go ahead. In that way the first crude mechanisms have developed into modern talking movies, radio, transatlantic telephones, telephotos, and television. More of the breed are awaiting birth in some expectant microscope or mathematical formula.
Group invention does not bring the spectacular rewards such as enriched lone prospectors like Pupin, Sperry, and Edison, for all patents are taken out in the name of the company, which expresses its pleasure in good work by increases in salary cheques. These salaries are not munificent—ten years after graduation a capable man will make, on an average, ten thousand dollars—but at least the workers are assured of some measure of security. Some of them resemble prosperous college department heads; some look like bank vice-presidents. None of them assume the airs of genius burning in secret.
Most of the Bell jugglers with electrons hide behind doors marked with red stars, which is the laboratory high-sign for “Do Not Disturb.” They will let you in these rooms, with affable guides who would frown if you tried to make drawings of things, but first they trot you around to the primary grade, beginning with the roof. On the sunny afternoon of my tour there was a sign out: “Rain Today,” warning roof visitors of artificial inclemency. Every day that nature doesn’t provide a shower, the Bell people make one themselves in order to pour water down on hundreds of small parts and speed up the effects of weathering. This is a fatigue test—to find out how sturdy an apparatus is under the rigors of outdoor life.
Mechanical fatigue tests are down below. There are rooms full of machinery whirring around for the impious purpose of wearing small moving parts to destruction. The dial on your telephone went through this inquisition, various models being twisted endlessly until the best model showed its superior stamina and was nominated for your fingers’ touch.
That test is an “investigation.” At the moment there are one thousand investigations going on at Bell Laboratories. Most of them are more complex than the fatigue tests, and some are miles above the layman’s head. In the hands of these searchers, problems which start out with the obviousness of high-school physics usually branch off into something on a rarefied intelligence level.
Noise, for instance, is a simple springboard into a deep pool of these intricacies. There are two acoustical-research rooms. The quiet one is insulated against outside sounds and proportioned against echo. If you stand still, you can hear your heart beat. Here a pin drops with the traditional clatter. Across the hall is a room designed to be noisy. A handclap or a yodel (popular with visitors) makes a fearful racket. Both rooms are used to study sound effects, and, quite incidentally, to send visitors home happy with the idea that they understand science.
But not far away is a red-star room where sound effects enter the field of modern magic. They hook a telephone transmitter onto the circuit of a vacuum tube. Speak into the transmitter and your voice makes jagged graphs in greenish light on a fluorescent screen at the end of the vacuum tube. Each vowel has a definite pattern. Your guide’s and your “E” are recognizable brothers. To make that effect your voice was transformed from physical sound waves in air to pulses of electric current, which, in turn, became millions of electrons beating against the fluorescent screen and forming visual images. It is a weird cycle—a sort of triple play from larynx to electron to optic nerve—and it brings together the research discoveries of scientists ranging from the father of electro-magnetics, Michael Faraday, to one of his disciples, G. W. Elmen.
Elmen is the guardian of this particular device. When it is not hooked up for visual-speaking, he uses it to test the useful magnetism of various metals. A non-magnetic subject produces a straight green line on the screen. A magnetic one makes a loop whose perpendicular length measures its magnetism. The most efficient performer is permalloy, which Elmen invented.
Soft Swedish iron used to be the best magnetic material. Elmen thought he could make a better magnet by combinations of alloys. He narrowed the components down to iron, cobalt, and nickel, and introduced the factors of degree of heat and rate of cooling. The thousands of possible combinations of these variables kept a staff in drudgery for years. There emerged permalloy, seventy-eight per cent nickel and twenty-two per cent iron, many hundred times more magnetic than iron alone. It is used to step up the message-sending capacity of cable and telephone wires. It does for present communication a service approaching in importance the introduction of underground conduits in the eighteen-eighties, when overhead wires were becoming impossibly numerous in all big cities and the device of burying them was necessary to further multiplication. Permalloy is one of those supports to our modern structure whose value cannot be rated in dollars. It was developed, not by an isolated genius in an attic alembic, but by a staff of precision workers, and as such it is typical of modern industrial research.
Most research, in fact, is endless patience geared to an objective. For thirty years hundreds of workers have been at the job of perfecting an automatic switchboard. They have an approximation of perfection now in the person of a robot they call suburban tandem. If it isn’t a person, it acts like one. Operator dials an out-of-town number for you. Suburban tandem—in appearance surpassing Rube Goldberg’s worst nightmare—stores up the dial pulses and decides whether the terminus is manual or automatic. If manual, it selects a suitable trunk line, flashes the operator on the other end, and, by means of a film voice-record stretched on a revolving drum, announces to her in human tones that New York wants Summit 0001. When something goes wrong with suburban tandem, it rings a bell for the repairman, and spends the waiting minutes in typing out a diagnosis to show him when and where its indigestion began. Suburban tandem, said our guide, is the smartest piece of remote control of our times.
Some research hasn’t anything to do with wheels or wires. There is a sanctum where mathematical sharks apply the laws of probability to the problems of wire installation. For instance, if every subscriber in the Ashland exchange wanted to talk to everyone in Rector, at the same moment, there would have to be some twenty thousand actual wires running between the two exchanges. But it would be unnecessarily expensive to provide for any such hypothetical peak. So figures are prepared showing how many people are likely to talk at any one time, whether it is a business or a residential section, whether the ladies of the district are inclined to be garrulous, how fast calls follow each other. To discover these telephone habits is like doping the chances of a crap game with unconventionally numbered dice. When the numerical data is gathered, the Bell mathematicians produce their tables of probability and figure out the least number of wires the two exchanges need to carry the greatest likely number of calls at any one moment. If they should underestimate, emergency calls can always be routed through other exchanges.
There is usually one major project just behind the Bell Laboratories, and another just ahead. A few years ago amateurs could catch occasional scraps of London broadcasts. Bell experts—primarily for telephone uses and incidentally for radio—set themselves the job of bringing sounds across the Atlantic with the clarity of local broadcasts. They spent two years, twenty-four hours a day, experimenting with Atlantic weather conditions. For a broadcast must be routed as carefully as a transatlantic flight. The sunlight creeping over the sea from Europe to America gives off a progressive wave of heat which affects the transmission level hour by hour. Winter and summer introduce seasonal variations. Storms bring chance factors. Distance (whether London or Berlin) must be considered. Bell engineers plotted every possible factor on charts, showing minute-by-minute conditions and what wave length to use at the moment. These charts are now filed at all transatlantic stations for the guidance of operators.
When King George first spoke to the United States, one amateur, who listened in on his own set, complained later that the royal words would suddenly fade into “How’s she going, Bill?” and similar impertinences. He complained in writing to the Bell people, who explained to him that he had remained on one channel, whereas the King’s speech had been shifted thirty times in order to use a wave of the length most efficient at crossing the Atlantic at the moment. The amateur had heard operators talking on the reserve channels.
The most recent achievement of the Laboratories is the improvement of airplane radiotelephony to the point where a pilot can be heard any time that he wants to talk to groundlings. For the past year two airplanes have been flying around New Jersey, by day and by night, in the worst weather they can find, near the ground and at high altitudes. A neat pattern of efficiency under all conditions was accumulated. The equipment engineers studied it and tinkered again with their devices. Already every mail pilot over the Alleghenies has a human voice from below to direct him.
Fairly soon the teletypewriter girl may make her Wall Street début. The machines are the same as those one hundred and twenty-two electrical typewriters now used in the New York State Police network. A message typed on one will reproduce on any, or all, of the others. The new trick is a switchboard for teletypewriters which will connect all subscribers to the service. You will call Operator by pounding “OPR.” She plugs in your correspondent and what you type on your machine comes out on his. This puts business deals in printed form, which is better confirmation than spoken words and cheaper than telegrams. If your party isn’t in his office, Operator lets you write on his teletypewriter and he finds your message when he gets back. Office wits will doubtless find possibilities for fun in this, if we may judge from experience with police teletypewriters. Recently an order went out telling the troopers not to tell each other amusing stories to kill time when their teletypewriters were not busy apprehending criminals.
Hollywood, of course, relies on Bell Laboratories for its technology. Eight months ago the Laboratories turned over to Hollywood an improved sound-recording system which nearly eliminates the buzzing noise which is continually present in all talkies made under the former method. To demonstrate the improvement they made a talkie which begins with the usual blurred sound background, then suddenly clears. The difference is impressive enough to make you conscious of a distinct gratitude for the release from noisy irritation.
There is still another talkie device up the Bell sleeve, this one designed to wash out the last remnant of extraneous sound. It is nothing more complex than being careful in the development of the films. The tap water now used is not clean enough; it must be filtered, heated to proper temperature, and fresh solutions made for each film. No more fine, free splashing around in the tub. Visually, the photoelectric eye is so much more critical than the human eye that the effect of cleaner development does not matter there; it is in scanning the sound track that impurities arise and offend the ear in reproduction.
Have no faith that three-dimensional movie heroines will soon appear on the screen with the rounded effect of your Uncle Stephen’s stereopticon collection of stage beauties. H. E. Ives, Bell electro-optics expert, has ascended the mountain and returns with scant hopes from the gods. In his office sits the picture of a smiling girl, her face minutely barred with lines. As you walk past, her eyes roll engagingly, her cheek turns, an unseen ear appears. This is a flat picture—not the two-faced changeling of advertising placards. It is authentic illusion; but it is not for the movie screen. According to Ives we must first develop apparatus “involving excessive speeds of operation, microscopic accuracy of film positioning, and photographic emulsions of speeds at present unknown.” He means by this, No.
Disc-recording is going to be better. Leopold Stokowski spent an afternoon at the Laboratories recently and listened to a recording by the “hill and dale” method, in which the stylus cuts up and down instead of waving from side to side, as formerly. Stokowski lent his ear attentively to an orchestra’s recording on the new device, but soon the “hill and dale” perfect recording began to provide what he considered imperfections on the part of the musicians. His habit of conducting got the better of him and he began to tear the unseen orchestra’s technique to pieces; he rearranged the seating, suggested shifting the position of the microphone, bawled out the French horns, and otherwise offered criticism to an insensate disc which could not hope to erase the grooves which fallible performers had marked upon it.
Besides its greater accuracy, which is due to a wider pitch range, the “hill and dale” has extremes of sound intensity one hundred times larger than old-style recordings. The new recording is now used for radio-broadcast libraries; in time it will be made practical for home phonographs.
Francis F. Lucas, head of the ultraviolet photomicroscope laboratory, has developed an instrument through which he sees things no eye has seen before. He magnifies objects six thousand times. The superlative tool was built to serve the ends of metallurgy; but with it Lucas also explores living tissue. Instead of dead cells pickled on a glass slide, he photographs living cells, in cross-sections so minute that a light-wave measurement must define the distance. In his pictures chromosomes take form, and biologists look with awe on the substance of their theories.
When the president of Bell Laboratories, Dr. Frank B. Jewett, began his career, physics and chemistry were considered fixed sciences. Engineers thought they could take data for granted. Since then a lot of disturbing ideas have evolved. Scientists have come to the conclusion that all chemical actions are electrical. They have decided that atoms combine because their electrons find more stable configurations. When man, then, begins to control electrons he becomes a manipulator of nature on a competitive basis with time and chance. The tricks he has so far performed on her are as a child’s experiments with a Christmas chemistry set. What he may do some day they will not talk about in West Street, but you carry away the impression that they wouldn’t be surprised at anything. ♦
Invention Factory
BY MALCOLM ROSS
TABLE OF CONTENTS
CREDITPHOTOGRAPH BY ALAMY / EVERETTThe Bell Telephone Laboratories, which is the research department of the American Telephone & Telegraph Company, is a good specimen of the modern invention factory, and handy to anyone who enjoys scientific peepshows. The ten-story main laboratory has fifteen acres of floor and takes up the block in West Street between Bethune and Bank Streets. Depression or not, the Telephone Company still finds it pays to spend nineteen million dollars a year there on research.
Except for the absence of elm trees and varsity letters, the Laboratories are something of a university, with eighteen hundred college graduates, seventy doctors of philosophy, and a crew of assistants, a total of five thousand people. Most of these are in the West Street building; but outposts are scattered from Colorado to Portnafrankagh, Ireland.
That seems an impressive outlay for a nickel in the Telephone Company’s slot. It appears, however, that not all of the nineteen million dollars goes immediately into improving the service. When a research worker gets on the side track of electronic physics, they tell him to go ahead. In that way the first crude mechanisms have developed into modern talking movies, radio, transatlantic telephones, telephotos, and television. More of the breed are awaiting birth in some expectant microscope or mathematical formula.
Group invention does not bring the spectacular rewards such as enriched lone prospectors like Pupin, Sperry, and Edison, for all patents are taken out in the name of the company, which expresses its pleasure in good work by increases in salary cheques. These salaries are not munificent—ten years after graduation a capable man will make, on an average, ten thousand dollars—but at least the workers are assured of some measure of security. Some of them resemble prosperous college department heads; some look like bank vice-presidents. None of them assume the airs of genius burning in secret.
Most of the Bell jugglers with electrons hide behind doors marked with red stars, which is the laboratory high-sign for “Do Not Disturb.” They will let you in these rooms, with affable guides who would frown if you tried to make drawings of things, but first they trot you around to the primary grade, beginning with the roof. On the sunny afternoon of my tour there was a sign out: “Rain Today,” warning roof visitors of artificial inclemency. Every day that nature doesn’t provide a shower, the Bell people make one themselves in order to pour water down on hundreds of small parts and speed up the effects of weathering. This is a fatigue test—to find out how sturdy an apparatus is under the rigors of outdoor life.
Mechanical fatigue tests are down below. There are rooms full of machinery whirring around for the impious purpose of wearing small moving parts to destruction. The dial on your telephone went through this inquisition, various models being twisted endlessly until the best model showed its superior stamina and was nominated for your fingers’ touch.
That test is an “investigation.” At the moment there are one thousand investigations going on at Bell Laboratories. Most of them are more complex than the fatigue tests, and some are miles above the layman’s head. In the hands of these searchers, problems which start out with the obviousness of high-school physics usually branch off into something on a rarefied intelligence level.
Noise, for instance, is a simple springboard into a deep pool of these intricacies. There are two acoustical-research rooms. The quiet one is insulated against outside sounds and proportioned against echo. If you stand still, you can hear your heart beat. Here a pin drops with the traditional clatter. Across the hall is a room designed to be noisy. A handclap or a yodel (popular with visitors) makes a fearful racket. Both rooms are used to study sound effects, and, quite incidentally, to send visitors home happy with the idea that they understand science.
But not far away is a red-star room where sound effects enter the field of modern magic. They hook a telephone transmitter onto the circuit of a vacuum tube. Speak into the transmitter and your voice makes jagged graphs in greenish light on a fluorescent screen at the end of the vacuum tube. Each vowel has a definite pattern. Your guide’s and your “E” are recognizable brothers. To make that effect your voice was transformed from physical sound waves in air to pulses of electric current, which, in turn, became millions of electrons beating against the fluorescent screen and forming visual images. It is a weird cycle—a sort of triple play from larynx to electron to optic nerve—and it brings together the research discoveries of scientists ranging from the father of electro-magnetics, Michael Faraday, to one of his disciples, G. W. Elmen.
Elmen is the guardian of this particular device. When it is not hooked up for visual-speaking, he uses it to test the useful magnetism of various metals. A non-magnetic subject produces a straight green line on the screen. A magnetic one makes a loop whose perpendicular length measures its magnetism. The most efficient performer is permalloy, which Elmen invented.
Soft Swedish iron used to be the best magnetic material. Elmen thought he could make a better magnet by combinations of alloys. He narrowed the components down to iron, cobalt, and nickel, and introduced the factors of degree of heat and rate of cooling. The thousands of possible combinations of these variables kept a staff in drudgery for years. There emerged permalloy, seventy-eight per cent nickel and twenty-two per cent iron, many hundred times more magnetic than iron alone. It is used to step up the message-sending capacity of cable and telephone wires. It does for present communication a service approaching in importance the introduction of underground conduits in the eighteen-eighties, when overhead wires were becoming impossibly numerous in all big cities and the device of burying them was necessary to further multiplication. Permalloy is one of those supports to our modern structure whose value cannot be rated in dollars. It was developed, not by an isolated genius in an attic alembic, but by a staff of precision workers, and as such it is typical of modern industrial research.
Most research, in fact, is endless patience geared to an objective. For thirty years hundreds of workers have been at the job of perfecting an automatic switchboard. They have an approximation of perfection now in the person of a robot they call suburban tandem. If it isn’t a person, it acts like one. Operator dials an out-of-town number for you. Suburban tandem—in appearance surpassing Rube Goldberg’s worst nightmare—stores up the dial pulses and decides whether the terminus is manual or automatic. If manual, it selects a suitable trunk line, flashes the operator on the other end, and, by means of a film voice-record stretched on a revolving drum, announces to her in human tones that New York wants Summit 0001. When something goes wrong with suburban tandem, it rings a bell for the repairman, and spends the waiting minutes in typing out a diagnosis to show him when and where its indigestion began. Suburban tandem, said our guide, is the smartest piece of remote control of our times.
Some research hasn’t anything to do with wheels or wires. There is a sanctum where mathematical sharks apply the laws of probability to the problems of wire installation. For instance, if every subscriber in the Ashland exchange wanted to talk to everyone in Rector, at the same moment, there would have to be some twenty thousand actual wires running between the two exchanges. But it would be unnecessarily expensive to provide for any such hypothetical peak. So figures are prepared showing how many people are likely to talk at any one time, whether it is a business or a residential section, whether the ladies of the district are inclined to be garrulous, how fast calls follow each other. To discover these telephone habits is like doping the chances of a crap game with unconventionally numbered dice. When the numerical data is gathered, the Bell mathematicians produce their tables of probability and figure out the least number of wires the two exchanges need to carry the greatest likely number of calls at any one moment. If they should underestimate, emergency calls can always be routed through other exchanges.
There is usually one major project just behind the Bell Laboratories, and another just ahead. A few years ago amateurs could catch occasional scraps of London broadcasts. Bell experts—primarily for telephone uses and incidentally for radio—set themselves the job of bringing sounds across the Atlantic with the clarity of local broadcasts. They spent two years, twenty-four hours a day, experimenting with Atlantic weather conditions. For a broadcast must be routed as carefully as a transatlantic flight. The sunlight creeping over the sea from Europe to America gives off a progressive wave of heat which affects the transmission level hour by hour. Winter and summer introduce seasonal variations. Storms bring chance factors. Distance (whether London or Berlin) must be considered. Bell engineers plotted every possible factor on charts, showing minute-by-minute conditions and what wave length to use at the moment. These charts are now filed at all transatlantic stations for the guidance of operators.
When King George first spoke to the United States, one amateur, who listened in on his own set, complained later that the royal words would suddenly fade into “How’s she going, Bill?” and similar impertinences. He complained in writing to the Bell people, who explained to him that he had remained on one channel, whereas the King’s speech had been shifted thirty times in order to use a wave of the length most efficient at crossing the Atlantic at the moment. The amateur had heard operators talking on the reserve channels.
The most recent achievement of the Laboratories is the improvement of airplane radiotelephony to the point where a pilot can be heard any time that he wants to talk to groundlings. For the past year two airplanes have been flying around New Jersey, by day and by night, in the worst weather they can find, near the ground and at high altitudes. A neat pattern of efficiency under all conditions was accumulated. The equipment engineers studied it and tinkered again with their devices. Already every mail pilot over the Alleghenies has a human voice from below to direct him.
Fairly soon the teletypewriter girl may make her Wall Street début. The machines are the same as those one hundred and twenty-two electrical typewriters now used in the New York State Police network. A message typed on one will reproduce on any, or all, of the others. The new trick is a switchboard for teletypewriters which will connect all subscribers to the service. You will call Operator by pounding “OPR.” She plugs in your correspondent and what you type on your machine comes out on his. This puts business deals in printed form, which is better confirmation than spoken words and cheaper than telegrams. If your party isn’t in his office, Operator lets you write on his teletypewriter and he finds your message when he gets back. Office wits will doubtless find possibilities for fun in this, if we may judge from experience with police teletypewriters. Recently an order went out telling the troopers not to tell each other amusing stories to kill time when their teletypewriters were not busy apprehending criminals.
Hollywood, of course, relies on Bell Laboratories for its technology. Eight months ago the Laboratories turned over to Hollywood an improved sound-recording system which nearly eliminates the buzzing noise which is continually present in all talkies made under the former method. To demonstrate the improvement they made a talkie which begins with the usual blurred sound background, then suddenly clears. The difference is impressive enough to make you conscious of a distinct gratitude for the release from noisy irritation.
There is still another talkie device up the Bell sleeve, this one designed to wash out the last remnant of extraneous sound. It is nothing more complex than being careful in the development of the films. The tap water now used is not clean enough; it must be filtered, heated to proper temperature, and fresh solutions made for each film. No more fine, free splashing around in the tub. Visually, the photoelectric eye is so much more critical than the human eye that the effect of cleaner development does not matter there; it is in scanning the sound track that impurities arise and offend the ear in reproduction.
Have no faith that three-dimensional movie heroines will soon appear on the screen with the rounded effect of your Uncle Stephen’s stereopticon collection of stage beauties. H. E. Ives, Bell electro-optics expert, has ascended the mountain and returns with scant hopes from the gods. In his office sits the picture of a smiling girl, her face minutely barred with lines. As you walk past, her eyes roll engagingly, her cheek turns, an unseen ear appears. This is a flat picture—not the two-faced changeling of advertising placards. It is authentic illusion; but it is not for the movie screen. According to Ives we must first develop apparatus “involving excessive speeds of operation, microscopic accuracy of film positioning, and photographic emulsions of speeds at present unknown.” He means by this, No.
Disc-recording is going to be better. Leopold Stokowski spent an afternoon at the Laboratories recently and listened to a recording by the “hill and dale” method, in which the stylus cuts up and down instead of waving from side to side, as formerly. Stokowski lent his ear attentively to an orchestra’s recording on the new device, but soon the “hill and dale” perfect recording began to provide what he considered imperfections on the part of the musicians. His habit of conducting got the better of him and he began to tear the unseen orchestra’s technique to pieces; he rearranged the seating, suggested shifting the position of the microphone, bawled out the French horns, and otherwise offered criticism to an insensate disc which could not hope to erase the grooves which fallible performers had marked upon it.
Besides its greater accuracy, which is due to a wider pitch range, the “hill and dale” has extremes of sound intensity one hundred times larger than old-style recordings. The new recording is now used for radio-broadcast libraries; in time it will be made practical for home phonographs.
Francis F. Lucas, head of the ultraviolet photomicroscope laboratory, has developed an instrument through which he sees things no eye has seen before. He magnifies objects six thousand times. The superlative tool was built to serve the ends of metallurgy; but with it Lucas also explores living tissue. Instead of dead cells pickled on a glass slide, he photographs living cells, in cross-sections so minute that a light-wave measurement must define the distance. In his pictures chromosomes take form, and biologists look with awe on the substance of their theories.
When the president of Bell Laboratories, Dr. Frank B. Jewett, began his career, physics and chemistry were considered fixed sciences. Engineers thought they could take data for granted. Since then a lot of disturbing ideas have evolved. Scientists have come to the conclusion that all chemical actions are electrical. They have decided that atoms combine because their electrons find more stable configurations. When man, then, begins to control electrons he becomes a manipulator of nature on a competitive basis with time and chance. The tricks he has so far performed on her are as a child’s experiments with a Christmas chemistry set. What he may do some day they will not talk about in West Street, but you carry away the impression that they wouldn’t be surprised at anything. ♦
沒有留言:
張貼留言