Opinion for the Court filed by MacKINNON, Circuit Judge.

In In re Net Weights for Determining Losses — Scrap Iron and Steel,1 the Interstate Commerce Commission (“Commission”), acting pursuant to the Administrative Procedure Act, 5 U.S.C. § 553(c) (1976),2 promulgated the following regulation to supplement its existing rules governing loss and damage claims by shippers against railroads:

Where weight is used as a measure of loss in rail transit of scrap iron and steel and actual tare and gross weights are determined at origin and destination, the settlement of claims shall be based upon a comparison of net weights at origin and destination.

The proceedings which resulted in the issuance of the above rule were initiated in response to a petition for investigation filed by Louis Padnos Iron & Metal Company, a shipper of scrap iron and steel and an intervenor in this case. The petition requested the Commission to examine the legality under section 20(11) of the Interstate Commerce Act (“Act”)3 of the practices of certain railroads in regard to settling claims for scrap iron and steel4 lost in transit. The practice in question involved the railroads determining the amount of scrap lost in transit by ascertaining the difference between the gross weights of the shipment at the points of origin and destination, and then reducing this amount by some percentage, varied in the judgment of the railroad 5 and allegedly designed to account for the inaccuracies of gross weight measurements, in arriving at a settlement figure.

The basis for Padnos’ objection to this method of settling shippers’ claims was that it did not comply with section 20(11) which mandates that common carriers “shall be liable . . . for the full actual loss, damage, or injury”6 to property incurred during the course of transit. The obvious objection to the railroads’ practice was that the use of gross weights reduced by some more or less arbitrary percentage was an unjustifiably inaccurate means of determining “full actual loss.” As discussed in somewhat greater detail below, where we consider whether the order here challenged was “arbitrary and capricious,” the intervenor made a strong case for the illegality of the railroads’ gross-weight calculations. *178Before the Commission, the railroads defended this practice in large part because the ability to vary the reduction factor gave them “greater flexibility” in dealing with claims.7 This insistence on “flexibility,” however, seems tacitly to concede that the settlements thus achieved did vary and were generally inaccurate, thus violating section 1(6) of the Act prohibiting “unjust . practice[s]” in rail rates, which by section 12(1)(a) the Commission is “required . to enforce . . . ”

The undisputed evidence before the Commission established that as scrap iron and steel are generally shipped in open-topped cars8 (“gondolas” in railroad parlance), the comparison of gross weights at origin and destination inevitably tends to be an imprecise measure of the “full actual loss” due to such sources of inaccuracy as precipitation and other forms of foreign matter potentially affecting the car’s weight.9 We are also mindful of a point apparent from the hearing record, but not pressed, that shippers may be the ones at fault when they load with clamshell loaders that scoop up bricks, concrete, plaster, wood, dirt, etc. along with the iron and steel, and then leave the debris in the car when they unload with crane operated magnets that remove only iron and steel. The operation of the rule, however, is a matter for the future.

Based on the statements presented by Pandos, the Association of American Railroads (AAR), the Institute of Scrap Iron and Steel, and others, the Commission determined that § 20(11) mandated the use of an accurate method for assessing the amount of loss.10 Obviously that method which eliminates as many extraneous factors as possible from weight comparisons at origin and destination would provide for the most accurate determination of “actual loss.” Thus, the net weights comparison set forth, ante, page 177 of 195 U.S.App.D.C., page 991 of 600 F.2d, was adopted since comparing gross weights eliminates none of the possible inaccuracies arising from such factors, and, at least in theory, a comparison of net weights would do so.

In fact, since net weights are determined not by direct weighing of the lading, but by subtracting tare (the weight of the car) from gross weight, a comparison of “net weights” does not eliminate all extraneous factors. For example, if the car containing the shipment is not completely unloaded at destination before the “tare” weight is determined, it will appear that there has been a net weight loss which actually is not the case. The Commission, however, was of the opinion that these potential inaccuracies were less significant than those affecting comparisons of gross weights.

Under the Commission rule, where actual tare and gross weights were taken at origin and destination, net weights must be used as the basis for determining actual loss. However, in response to the protestations of the Association of American Railroads the Commission did permit the railroads to use a reasonable “tolerance factor” to adjust net weight figures because of discrepancies between scales at origin and destination,11 and also allowed the carrier the option of obtaining a third weighing at the destination to corroborate the accuracy of the initial net loss determination. In the same order in which the challenged regulation was established, the Commission also instructed the railroads to cease and desist from using or applying their current methods for settling damage and loss claims to *179the extent that these practices were inconsistent with the net-weight regulation.12

The AAR seeks to invalidate the ICC’s regulation on the grounds that its promulgation was not within the Commission’s authority, and that even if enacting the regulation was a valid exercise of the ICC’s jurisdiction, the content of the rule is unreasonable. We find both these arguments without merit and affirm the Commission’s order in its entirety.

I

THE COMMISSION’S AUTHORITY

The net weights regulation involved here applies only to voluntary settlement of loss and damage claims. It does not apply when such claims are litigated in court. If a railroad does not want to settle a claim in accordance with the net weight regulation, it is free to take the matter to court, where it can offer into evidence all relevant facts. Thus, in this case, the Commission merely seeks court recognition of the rather narrow authority to promulgate regulations governing voluntary settlements of loss and damage claims.

In support of the Commission’s position that it had the requisite authority to promulgate the net weight regulation, two different rationales are advanced. First, the Commission relies upon its recent decision in Ex Parte No. 263, Loss and Damage Claims, 340 I.C.C. 515 (1972) [hereinafter Loss and Damage Claims].13 In that proceeding the Commission asserted, without challenge from the railroads, a jurisdiction to promulgate regulations implementing the Act provided that such regulations did not interfere with the actual determination of the merits or the measure of damages in particular loss and damage claims, which the ICC conceded were questions cognizable only in the courts.14 Petitioners acknowledge the validity of the rule-making jurisdiction over settlements claimed in Loss and Damage Claims.15 They formulate their argument around the issue of whether the net-weight regulation is more properly characterized as “substantive” or “procedural.” The Commission acknowledged that substantive regulations are beyond the scope of its jurisdiction as asserted in Loss and Damage Claims, but contends it is authorized to issue procedural regulations governing settlements.

A second approach, elaborated at length by intervenor Padnos, ignores the questions of the substantive or procedural nature of the net-weight rule and relies on a more indeterminately based general rule-making power vested in the Commission to justify its jurisdiction.16 In Loss and Damage Claims, the ICC, in addition to developing the substantive/procedural dichotomy mentioned above, alluded to this second rationale for maintaining its rule-making authority in this case:

The Congress, recognizing that urgent and complex matters are involved in attempting to regulate the interstate transportation of passengers and property, and realizing also that numerous unforeseen problems could arise in the course of such regulation, has entrusted to us, in section 12(1) of the Act and cognate provisions in parts II, III, and IV thereof, broad regu*180latory power to administer and execute the provisions of the statute and to carry out the congressionally declared national transportation policy. [49 U.S.C. preceding §§ 1, 301, 901, and 1001],

340 I.C.C. at 542-43.

Under this analysis, although section 12(1) does not contain an express grant of rulemaking authority,17 it is argued that the fact that sections in other parts of the Act to which it is “functionally equivalent” have been held to convey such authority (it being expressly stated, e.g., 49 U.S.C. §§ 304(a)(6), 904(a), 1003(a)), implies that a similar authority should be inferred to exist in the context of 12(1).18 See American Trucking Ass’ns v. United States, 344 U.S. 298, 311-312, 73 S.Ct. 307, 97 L.Ed. 337 (1953). Furthermore, the general principle that when an agency is entrusted with the supervision of an industry its power to issue rules necessary to do so should be broadly construed, American Trucking Ass’ns, supra, 344 U.S. at 309-312, 73 S.Ct. 307; General Tank Car Corp. v. El Dorado Terminal Co., 308 U.S. 422, 60 S.Ct. 325, 84 L.Ed. 361 (1940), is asserted in support of the Commission’s authority to promulgate the net weight rule in exercise of the statutory mandates of section 1(6)19 and 20(11).

*181 Although the second argument advanced in support of the Commission’s authority to require the use of net weights is not without force, our conclusion that the regulation in question is plainly procedural makes it unnecessary to reach the second issue.20 By statute: “The Commission is authorized and required to execute and enforce the provisions of [that] . part [of the Interstate Commerce Act applicable to carriers by rail].” 49 U.S.C. § 12(l)(a). This includes the declared “. . . national transportation policy of the Congress to provide for the fair and impartial regulation of all modes of transportation subject to the provisions of the [Interstate Commerce] Act, so administered as [inter alia] . . . to encourage the establishment and maintenance of reasonable charges for transportation charges without unjust discrimination, undue preferences or advantages . . . . All of the provisions of this Act shall be administered and enforced with a view to carrying out the above declarations of policy.” Act of September 18, 1940, c. 722, Title I, § 1, 54 Stat. 899. 49 U.S.C. preceding § 1 (1970). (Emphasis added).

In the course of the present rule making proceeding, the Commission found instances of unjust discrimination, undue preferences and advantages in the railroads’ settlement procedures. This discrimination resulted from the statutory liability of “full actual loss,” not being paid in a great many instances. Instead, “[settlement offers of 25, 50, or 75 percent of claims appealed] to be based more on the bargaining position of the parties than the liability of the carrier for “full actual loss” under § 20(11), 352 I.C.C. 413. In other instances the amount of loss for which the carrier was responsible was not determined accurately. Id., 410. In sum, the Commission found that the “flexibility” in claims’ settlements, that the railroads acknowledge and support, was an improper endeavor arbitrarily and preferentially to limit claims liability. Id., 410.

The net weight regulation in effect prescribes an evidentiary rule that net weight, whenever it is available, is to be used in all settlements in preference to gross weight figures to determine “full actual loss” based on weight. It is true that some evidentiary rules may be substantive, but in our judgment the rule here is not. In general, substantive law is that which creates underlying rights and duties,21 while procedural law includes that which is embraced by evidence as well as pleading and practice.22 In the instant case, the substantive law embodied in the Interstate Commerce Act declares, creates, and defines shippers’ right to recover their “full actual loss,” and the corresponding duty of the railroad to compensate all shippers to that extent, 49 U.S.C. § 20(11). The net weight regulation, an endeavor by the Commission to carry out the National Transportation Policy supra, and the responsibility imposed by the requirement of § 12(1) of the Act “to execute and enforce the provision” mandating the payment of “full actual loss,” § 20(11), does not add any substantive liability to the railroads’ obligation. Rather, *182this regulation merely prescribes the evidentiary method that the Commission, after full proceedings, found best suited to determine “full actual loss” for settlement purposes.23 Theoretically, using net weights or gross weights to determine loss should produce precisely the same result, certainly one method is not substantively different from the other. The Commission has simply determined that mechanically the former is more accurate. Far from altering the railroad’s substantive liability, the net weight regulation does no more than require the extent of this liability to be determined by the best information available. The regulation is thus basically procedural.

Not only do the petitioners concede that the regulation is authorized if it is procedural,24 but also, given Congress’ broad delegation to the ICC of the responsibility for the enforcement of the National Transportation Policy, supra; e.g., Board of Trade of Kansas City, Mo. v. United States, 314 U.S. 534, 62 S.Ct. 366, 86 L.Ed. 432 (1942), we would not adopt a narrow view of the Commission’s authority to establish regulations so intimately connected to the discharge of its statutory responsibilities as the net weight regulation is to the determination of the amount of “full actual loss.”

In Secretary of Agriculture v. United States, 350 U.S. 162, 76 S.Ct. 244, 100 L.Ed. 173 (1956), the Commission, after finding with respect to claims for eggs damaged in transit, that estimates of higher-percentage pre-carrier damage were incorrect, by order authorized the railroads to include in their tariff schedules the following tolerance provision:

On eggs placed in packages at rail point of origin of the shipment, no claim shall be allowed where the physical damage to the egg at destination does not exceed 3% of the contents of the packages containing damaged eggs. Where damage exceeds 3%, claims shall be allowed for all damage in excess of 3%, if investigation develops carrier liability.

Exception. — Where bona fide certificates of Federal or State egg inspection agencies showing extent of physical damage to eggs determined at rail point of origin of the shipment immediately prior to tender for rail transportation indicate the actual shell damage to be other than 2%, the percentage of actual damage as shown on such certificates, plus 1% shall be used in lieu of 3% specified in this Section.

350 U.S. at 164, 76 S.Ct. at 246. This order amounts to a regulation of the railroads’ handling of such claims and is not substantially different from the rule here applicable to claims for scrap iron and steel allegedly lost in transit. In the case of egg claims the Commission authorized a tolerance of 3% for eggs packaged at rail point of origin; with respect to scrap iron and steel claims the Commission provided that where net weights are determined at origin and destination the settlement of loss claims must be based thereon. In both instances, the Commission regulated the payment of claims. In the egg case, the Supreme Court did not question the authority of the Commission to take the action it did.25 We similarly find no basis for ques*183tioning the Commission’s authority in this case.

We also note, but need not rely on, the following observations. The Interstate Commerce Act does not specify all of the multitudinous facets of carrier transportation that are to be administered under the Act. It leaves room for the Commission to exercise its discretion and because of that vested discretion, it may be forcefully argued that the Commission is empowered to promulgate such reasonable rules and orders it finds to be necessary to carry out its statutory duties, even without an explicit grant of rulemaking power. Professor K. Davis in his Administrative Law Treatise (Supp. 1970) states the general rule to be:

Any officer who has discretionary power necessarily also has the power to state publicly the manner in which he will exercise it, and any such public statement can be adopted through a rule-making procedure, whether or not the legislative body has separately conferred a rule-making power on the officer.

§ 6.16, p. 285. See, e.g., Electrolux Corporation v. Miller, 286 N.Y. 390, 393, 36 N.E.2d 633, 636 (1941). See also, British Caledonian Airways, Ltd. v. CAB, 190 U.S.App.D.C. 1, 584 F.2d 982 (1978).

More recent writers in the field have described in even broader terms the implicit power of agencies to issue such rules, substantive or procedural, as may be necessary to the effective exercise of their authority:

all administrative agencies, whether regulatory or law enforcing, except where an express statutory provision is to the contrary, have substantive rule making power within the limits of their statutory authority.

Mezines, Stein, Gruff, Administrative Law § 16.03[1] at 16-25; cf. National Petroleum Refiners Ass’n v. FTC, 157 U.S.App.D.C. 83, 482 F.2d 672 (1973), cert. denied, 415 U.S. 951, 94 S.Ct. 1475, 39 L.Ed.2d 567 (1974); Wright, The Courts and the Rule Making Process: The Limits of Judicial Review, 59 Corn.L.Rev. 375-377 (1974). Sound logic supports these observations, but we rely on the previously stated procedural grounds in holding that the net weight regulation to be applied in settlements was justifiably promulgated by the Commission in discharge of its statutory obligation to enforce the obligation of railroads to compensate shippers for their full actual loss, 49 U.S.C. §§ 12(a), 20(11).

II

THE NET WEIGHT REGULATION IS NOT ARBITRARY OR CAPRICIOUS

Petitioners’ argument that the net-weight regulation is arbitrary and capricious because net weights are no more, and may be less accurate, than gross weights is easily answered. The scope of review by this court of orders by the Commission is narrow, and even if we might question whether gross weights tended to be more accurate than net weights, we are not at liberty to substitute our judgment for that of the ICC on that matter. United States v. Pierce Auto Freight Lines, Inc., 327 U.S. 515, 535-536, 66 S.Ct. 687, 90 L.Ed. 821 (1946); Rochester Telephone Corp. v. United States, 307 U.S. 125, 138-140, 59 S.Ct. 754, 83 L.Ed. 1147 (1939); Ethyl Corp. v. EPA, 176 U.S.App.D.C. 373, 405-410, 541 F.2d 1, 33-38 (1975), cert. denied, 426 U.S. 941, 96 S.Ct. 2663, 49 L.Ed.2d 394 (1976); National Ass’n of Food Chains, Inc. v. ICC, 175 U.S.App.D.C. 346, 351-352, 535 F.2d 1308, 1313-1314 (1976). Moreover, we find abundant support for the position the Com*184mission has taken. Even if the judgment of the Commission were not entitled to deference we would be inclined to agree that net weight measurements tend to be a more accurate method of determining “full actual loss” than are gross weight figures.

During the hearing before the Commission,26 all the evidence indicates that due to precipitation, rubbish, etc. gross weights were inaccurate measures of losses in transit.27 These extraneous factors can be largely eliminated through the use of net weight rather than gross weight figures, and this fact alone would justify the Commission’s order. Moreover, petitioners’ own practices do not support their attempt to discredit the accuracy of net-weight figures, as it appears that the railroads themselves use net weight to determine freight charges,28 Thus, the Commission’s position not only has the strength of common sense, but also is supported by the practice of the railroads themselves. One can assume that the railroads desire an accurate reading of the amount of freight they are carrying.29 We accordingly conclude that petitioners have not discharged the heavy burden of proving that the Commission in requiring the use of net weights had acted arbitrarily and capriciously.

A large part of the evidence presented by the petitioners before the Commission consisted of an enumeration of the hardships that would result from requiring the use of net weight in the determination of loss settlements.30 That the AAR would attempt to support the use of gross weight by such policy arguments suggests that it also recognized the difficulty of arguing in favor of gross weight figures on the basis that they allowed more reliable calculation of the statutorily mandated “full actual cost” than net weight figures.

Petitioners’ main argument before this court is that in transporting scrap iron and steel the lading itself is never weighed independently. Rather, the loaded and tare weights of the cars are determined. Thus, mandating the use of net weights requires four weighing — gross and tare at origin and destination — instead of only two — gross only at origin and destination. The AAR argues that by increasing the number of required weighings from two to four, the Commission’s regulation simply interjects another possible source of error into the assessment process.31 The ICC took consequence of the possibility of scale inaccuracies by allowing for a “reasonable tolerance” for variations in net weight.32 Petitioners’ point in this respect, however, is not totally frivolous. On the other hand, one would expect inaccuracies to arise not so much in discrepancies between different weighings made on the same scale, but in discrepancies between weighings made on different scales. It is thus hard to see how in practice the net weight rule would create a significantly greater potentiality for error than the use of gross weights33 since differ*185ent scales are also used for determining gross weights. What the AAR overlooks is that the major eradicable source of error in weighing loads carried in gondolas is the possibility of foreign matter being either deposited or removed from the cars between weighings. We cannot say that it was irrational of the Commission to consider that removing this source of error merited incurring the slight additional risk of mistake involved in requiring four instead of two weighings. Moreover, the Commission, through its allowance of a “reasonable tolerance” and the option of a third weighing has minimized the effect of this latter risk, while a gross weight system does nothing to reduce the possible error derived from the injection of foreign matter.

There is no allegation here that the Commission violated any principle of due process in the conduct of its hearing, and the petitioners were allowed ample opportunity to present their case; the Commission has not acted arbitrarily and capriciously simply because it was not convinced by their presentation. We dismiss the petition for review and affirm the Commission’s order in its entirety.34

So Ordered.