On October 1,1907, an act of the General Court (St. 1907, c. 535) went into effect entitled “ An act to provide for the testing and sealing of weights, measures and balances having a device for indicating the price as well as the weight of commodities offered for sale.” By § 1 it is provided that R. L. c. 62, relating to testing and sealing weights, measures and balances “ shall apply to all scales, balances, computing scales and other devices having a device for indicating or registering the price as well as the weight of the commodity offered for sale. All such computing devices shall be tested as to the correctness of both weights and values indicated by them.” By § 2 it is provided that “ a sealer or deputy sealer shall seal such devices when tested and found correct, or shall mark, condemn or seize such devices if incorrect,” in accordance with R. L. c. 62 ; and all penalties imposed for violation of the provisions of R. L. c. 62, relative to weights, measures and balances shall apply to such devices.
*505On November 29, 1907, this bill in equity was brought by the plaintiff, an Illinois corporation, against the sealer of weights and measures of the city of Cambridge. The bill in substance alleges that it is the sole vendor of weighing and value indicating scales manufactured by the Dayton Company of Ohio, known as barrel shaped computing scales and manufactured on what is called the secondary principle; that the weights and values indicated by them are correct and accurate; that it has sold large numbers of them in Cambridge, in Massachusetts and in the United States generally, on the conditional sale instalment plan; and that money is still due the plaintiff on many of these sales. The bill further alleges that the defendant intends to condemn all secondary principle barrel shaped computing scales, and seeks to have him enjoined from so doing. The defendant filed an answer which in substance denied that the aforesaid scales were correct and accurate in the values therein set forth. Issue was joined. The case was heard on the merits, the evidence was taken by a commissioner, a decree was entered dismissing the bill, and the cause is now before us on an appeal from that decree.
The plaintiff and the defendant have argued at length the correctness of the kind of computing scales here in question. But the correctness of these scales is not now before us. The decision of that question is committed by the statute to the sealers of weights and measures of the several cities and towns of the Commonwealth. If a sealer of weights and measures in deciding that question should proceed on erroneous principles of law, his decision could be quashed on certiorari, and in some cases a writ of mandamus might issue directing him to take specific action. See for example Darrigan v. Williams, 198 Mass. 457. But there is no jurisdiction in this court to take from the public officer selected by the act the duty of deciding the question committed to him for decision by the act, on the ground that he threatens to come to a wrong conclusion. That is what this plaintiff asks the court to do so far as the correctness of the threatened decision by the. defendant (as sealer of weights and measures for the city of Cambridge) is concerned.
There is however a second ground on which the plaintiff seeks to maintain this bill in equity, namely, that St. 1907, c. 535, is *506unconstitutional. That question is properly raised by a bill in which the plaintiff seeks to have the defendant enjoined from taking action under it which will affect his rights of property. Board of Liquidation v. McComb, 92 U. S. 531. Such a bill proceeds on the same footing as a bill to enjoin a public officer from doing an act colore officii which is in fact and in law outside his jurisdiction. See for example Owens v. Crossett, 105 Ill. 354; Wetherell v. Newington, 54 Conn. 67; Morgan v. Miller, 59 Iowa, 481; Belknap v. Belknap, 2 Johns. (Ch.) 463.
In passing upon this second question the arguments put forward on the first question are pertinent although that question is not now directly before us. No better way can be found for demonstrating what the statute has committed to the final decision of the three hundred and fifty-four sealers of weights and measures of the cities and towns of the Commonwealth than to state the arguments, which have been made here on the question which those officers have to decide. We speak of the question which has been finally committed by St. 1907, c. 535, to the three hundred and fifty-four sealers of weights and measures because there are thirty-three cities and three hundred and twenty-one towns in the State.
Before taking up the arguments put forward here upon the correctness of these computing scales, it is necessary to understand what they are. As we have said, the scales made by the plaintiff are self-computing scales, built on what has been styled the secondary principle. Scale No. 63 was taken at the trial as a type of these scales. The chart indicating the value of the several weights is the thing on which the correctness or incorrectness of this scale depends, so far as it registers the prices.
A facsimile, somewhat reduced in size, of portions of this chart is printed on the opposite page.
The centre is duplicated in the portions given because the limits of a single page do not admit of otherwise reproducing a facsimile.
The large figures, 23 — 0 — 1, reading down on the left of the centre, and 13—12 —11, reading up on the right of it represent the pounds, and the intermediate figures 4, 8 and 12, the ounces, which the article put upon the scale weighs. The figures in the several columns in alignment with the one-pound line, *507for example, represent the value of one pound at the prices there stated. We assume (although it is not so stated in the evidence) that the first of the two figures 13 means 12^.
This chart is put around a small barrel or drum which contains a set of balances and springs which weigh the article, the
weight and value of which are to be stated by the scale in question.
For example, if the article weighs 23 pounds 4 ounces, and is 14 cents a pound, the operator finds the value by looking at the figures in the 14 cent a pound column opposite that weight, and he finds it to be 326 cents.
*508The drum on which this chart is rolled is under a cover. That cover, on the side of it next to the dealer, has a slit extending across the whole chart, to enable him to read the value as well as the weight of the article put upon the scale. There is also a slit on the customer’s side of the drum, which enables him to see the centre of the chart, that is, the weight and the weight only. The reason why the figures on the two sides of the centre are in inverse order is to enable the same figures to appear on the opposite sides of the chart at one and the same time.
Scale 63, in the words of the plaintiff’s principal witness, is intended “ for the butcher and market trade.” As will be seen on inspection, it indicates each ounce of the meat or other piece of provisions sold, and the value of each two ounces at the several prices for which the chart is drawn up, when meat or other provisions are sold by weight. The evidence showed that scale 63 and a similar scale numbered 61 are also used, and are intended by the plaintiff to be used, to indicate the correct weight when so many cents’ worth of meat or other provisions at a specified price are asked for by the customer and sold by the dealer. This evidence also showed that among the poorer classes this method of dealing is not an infrequent one.
The numbers given in the centre column confessedly state with correctness the weight of the article on the scale. As a scale the machine is correct.
The several numbers given in alignment with every two ounces are confessedly commercially correct values where an article is sold by weight. The plaintiff contends that they are correct within the meaning of St. 1907, c. 535. The defendant contends that the values indicated for the odd ounces are in many- cases not even commercially correct. What is meant by these two contentions- is best shown by an example. Suppose a customer asks for half a pound of cheese at 11 cents a pound. A piece is cut off a cheese and placed upon the scale, and the indicator in the centre shows that it weighs 9 ounces. The same indicator will be exactly half way between two lines, one in alignment with 8 ounces and the other in alignment with 10 ounces.
0 In the case put it will be exactly between the figures ¡j!. The *509price indicated by the chart is half way between 6 and 7 cents, that is to say, 6J cents. The half cent cannot be paid in addition to 6 cents nor returned if the customer pays 7 cents. Under the commercial custom, which is the plaintiff’s sole justification for the figure 6 on the chart now in question and for similar figures placed upon it (as we shall see hereafter more at length), the dealer takes the half cent and charges the customer 7 cents for his cheese.
In point of fact the value of the cheese in the case put is 6^-cents, and under the commercial custom the fraction being less than half a cent belonged to the customer, not to the dealer. The amount legally due for the piece of cheese in question is 6 cents.
The reason why the chart (in the case put above) gives ££ of a cent to the dealer in place of giving of a cent to the customer, is that a fraction of a cent in favor of the dealer is included in the figure 6 opposite 8 ounces. Eight ounces at 11 cents a pound are 5-¡- cents. The chart gives the dealer the benefit of the half cent and calls eight ounces 6 cents. Therefore, when the dealer taking the mean between 6 cents and 7 cents for 9 ounces finds that the price is 6J cents, and takes to himself that half cent, he gets a fraction of a cent twice.
The plaintiff’s first answer to this is that this is a dishonest use of an honest chart, that it is dishonest to use a commercially correct figure as an arithmetically correct figure, and that if that be done it is not the chart that is to blame but the dealer who uses it dishonestly.
That argument assumes that the ordinary dealer knows that some of the prices on chart 63 are commercially correct and some arithmetically correct. More than that, that argument assumes that the ordinary dealer knows which of the 4,608 prices on chart 63 are commercially and which are arithmetically" correct. Still further, that argument assumes that the ordinary dealer in using the chart can 'and will in the hurry of business carry in mind which of the prices are commercially and which are arithmetically correct, and, in ascertaining the price not stated in terms on the chart but indicated upon it by the marker coming between two of the prices placed thereon, will bear in mind which prices are commercially and which are arithmetically *510correct, and take to himself or give to his customer the benefit of the fractions accordingly.
There is nothing on this chart that tells the dealer who buys it that some of the prices are not arithmetically correct. In the absence of such a notice the ordinary dealer would not even necessarily know that any of them are not arithmetically correct. But when the other two assumptions underlying this argument are reached, namely, that some prices on it are only commercially correct and the dealer using it should not only know which are arithmetically and which are commercially correct, but should act upon that in the hurry of business in giving or taking the fractions when fractions are involved in the value of odd ounces, it may well be urged that the argument cannot carry the load put upon it and collapses.
The next answer made by the plaintiff’s counsel is that the dealer has no right to use the chart to get the value of the odd ounces.
Apart from special directions from the plaintiff, it might well be thought that when the indicator points to an odd ounce on this chart, exactly half way between the even ounces, it indicates that the value of the odd ounce is exactly half way between the values given for the even ounces, and that the defendant is chargeable with knowledge that that is the meaning of the chart in such a case, and is chargeable also with knowledge that dealers will so understand and use it.
But this was not the only answer made to this argument of the plaintiff. The other answer made was that the plaintiff has told purchasers of its machines to do the very thing which it now says ought not to be done, and which, if done, it now says is wrong, and which (it now says) is the wrong of the dealer and not its (the plaintiff’s) wrong. There was evidence that on June 14,1906, the plaintiff sent to its salesmen two stickers to be put “ just above the opening so it can be readily seen by the operator of the scale,” “ wherever you find one of our barrel-shaped scales.” The first sticker to be used on scales where an indicator records the weight in the centre of the chart is as follows: “ To ascertain value, read figures adjacent to the price per lb when indicator is on even ounces; when indicator is on odd ounces or between even ounces take the mean of figures.” *511The second sticker to be used where an indicating line is used in place of the indicator is: “ For value, read figures when indicating line is directly over center. When indicating line is between center, take mean of figures.”
This brings us to the contentions of the plaintiff and of the defendant as to the other use made of computing scales Nos. 63 «-and 61, namely, where a customer asks for so many cents’ worth at a specified price and the scale is used to ascertain what weight should be given.
The alleged incorrectness of the chart when used in this way is also most easily shown by an example. Suppose four cents worth of cheese is asked for at seven cents a pound. In the seven cents a pound column there are two 4’s, one opposite 8 ounces and the other opposite 10 ounces. A dealer naturally would or certainly might take the upper of these two four cents and deliver to his customer 8 ounces. The customer is entitled to f of a pound, or a little more than nine ounces.
It is urged that the incorrectness of this chart when used in this way (to find the weight from the price) comes from the fact that values which are commercially but not arithmetically correct have no place upon such a chart. It is urged still further that they have no place on such a chart for these reasons: Where a customer buying by weight asks for an amount the price of which is not divisible by our currency, the fraction of a cent must be given to the customer or taken to himself by the dealer. But where the amount asked for by the customer is to be measured by its value and the dealer consents to sell on that basis, he is bound to give the customer the exact amount due him at that price if it can be weighed out to him.
In the case put above the customer is entitled to 9 ounces and % of an ounce. It may be that there is a question as to how this | of an ounce is to be dealt with in case of a scale which gives only ounces, for example. How such a fraction should be dealt with in such a case need not be determined here, it is urged. But what is clear as matter of law (it is also urged) is that the dealer in the case put above does not give the weight due to the customer when he gives the customer 8 in place of 9 ounces. The fraction to be disregarded here (if there be one) is the fraction of the ounce, not the fraction of a cent. And prices which *512are only commercially (not arithmetically) correct because calculated to give to one or the other those fractions of a cent which result from a sale by weight are not correct when used to sell by the money’s worth and not by weight.
The first answer made to this by the plaintiff’s counsel was that the chart was designed to give value for weight, and if to give weight for value it is used in a way for which the plaintiff is not responsible.
Here again it is a matter of common experience that in the “ butcher and market trade ” articles are bought and sold on the basis of so many cents’ worth at a specified price. And there was evidence in the case at bar confirming this matter of common experience. In selling such an automatic self-computing scale under these circumstances the plaintiff must be taken to have intended it to be used not only to show the price of a specified weight but the weight at a specified price.
More than that, in making this contention there was evidence that the plaintiff for a second time was not altogether frank. In a circular sent out by the plaintiff to dealers who had bought No. 61 scales, to be distributed by them among their customers, is found this statement: “ The use of this machine renders it impossible for you to lose one cent on account of short weight or a mistake in calculation. For example: Should you call for twenty-five cents worth of an article, when that article is placed upon the machine it will show its exact value in money.” An attempt was made at the trial to confine this circular to scales of another type. But there was evidence that at the head of this circular there was the representation of a scale which is in fact a cut of a No. 61 scale.
We come now to the argument that St. 1907, c. 535, is unconstitutional.
The first contention made by the plaintiff’s counsel in this connection is that since by the construction of these scales (taking No. 63 as a type) the scale is not the arbiter of the price but only of the weight, the regulation of the computing part of such a scale is not within the police power of the State. His argument in support of that contention is, to quote his own words, “ In ascertaining the value of a given weight at a certain price per pound, there is no common arbiter in the control of *513either party to which both parties appeal. The sole arbiter is the multiplication table whose dictates are changeless. But the mental process of computation of either purchaser or dealer, the other party to the transaction does not accept. He makes it for himself. Hence any instrument making a computation for the dealer does not enter into the transaction; just as the dealer’s table of values would not enter into a sale, so the computed values on the chart of the computing scale do not enter into the transaction over the computing scale. The law has no more right to interfere with the chart of computed values in the scale than it would have to interfere with the table of computed values kept by a merchant.” 1
We have no doubt that it would be within the police power of a State to forbid the sale of incorrect tables of value to be used by dealers as correct tables in making sales of articles dealt in by them over an ordinary pound and ounce scale. In saying this we mean to put the case where the table is to be hung up on the dealer’s wall and is not to be seen by the customer. By such a table of values not only would an honest dealer be tricked into making dishonest charges, but at least many customers would be cheated. For it is a matter of common experience that many customers accept the computation made by the dealer without verifying its correctness.
We are also of opinion that it is equally within the police power of the State to forbid the sale of such a table when it is made part of an automatic self-computing scale.
Had the Legislature enacted a law forbidding the sale of self-computing scales like the plaintiff’s scale No. 63, it would (in our opinion) have been within the police power of the General Court. But that is not what the Legislature did in enacting St. 1907, c. 535.
More than that, in St. 1907, c. 535, the Legislature did not enact a statute where authority is given to an administrative officer to forbid the sale of a thing which can be regulated under the police power of the Legislature, in case he deems it to be contrary to law, until the court can determine the legal questions involved, as was the case in New York Ins. Co. v. Hardison, ante, 190. The act here in question (St. 1907, c. 535) makes the decision of each one of the three hundred and fifty-four *514sealers of weights and measures of the Commonwealth final upon the correctness of “computing scales and other devices having a device for indicating or registering the price as well as the weight of the commodity offered for sale.” The statute provides that: “All such computing devices shall be tested as to the correctness of both weights and values indicated bv them.”
If the correctness of the values indicated by such scales and other devices is to be tested by arithmetic, that is to say, if the values are to be arithmetically correct, the acts of the several sealers of weights and measures are of the same character as their acts in determining the correctness of an ordinary pound and ounce scale and the statute is constitutional. The act of a sealer of weights and measures in determining whether a scale balances when nothing is on it and whether it indicates accurately the standard weights when they are placed upon it is an act ministerial in its character. All that the officer is called upon to do in performing this act is to be correct in his observation. Such an act not only may be but must be left to an executive officer. The determination of the correctness of a table of weights and values which is required to be arithmetically correct is of the same character.
But if the correctness to be finally passed upon by a sealer of weights and measures under St. 1907, c. 535, is the commercial correctness of the values indicated by such scales or other devices as the correct value for a given weight, that is to say, the sum legally due for a certain weight, and conversely, when articles are sold by the cent’s worth, the weight legally due for the value specified, we are of opinion that the act would be unconstitutional. Such an act would leave to the final determination of a sealer of weights and measures the questions discussed in the arguments made at the bar as to the correctness of the computing scales under consideration here. That is to say, it would leave to an executive officer (first) the determination of the question how great a departure from mathematical accuracy would be permissible in determining the values of small articles sold by weight involving fractions of a cent without making the transactions commercially and therefore legally incorrect, and (second) the question whether a table of values *515which are only commercially correct because they disregard fractions of a cent is the proper basis for a table to be used to find the true weight of articles sold by the cent’s worth. Those questions are judicial questions and cannot be left to the final determination of an executive officer.
J. M. Zane (of Illinois), (W. H. Coolidge & C. F. Morse with him,) for the plaintiff. G. A. A. Pevey, (A. H. Meads of Illinois & S. J. Elder with him,) for the defendant.In addition we are of opinion that the Legislature did not intend to leave to the final determination of three hundred and fifty-four municipal officers the final determination for the territory covered by their several municipalities the puzzling questions upon which the commercial correctness of these computing scales depends. It is enough to point out in that connection that different results might be expected to be reached in this question by such officers.
For these reasons we are of opinion that the General Court when it enacted St. 1907, c. 535, by which the correctness of self-computing scales and the other devices therein mentioned was committed to the final decision of the sealers of weights and measures for the several cities and,towns of the Commonwealth, must (in our opinion) have intended that the values to be placed on such charts should be arithmetically correct. So construed, St. 1907, c. 535, is a valid statute, and the decree dismissing the bill must be affirmed.
So ordered.