MANTON, Circuit Judge.

This libel was filed by the assignee of the Fruit & Produce Exchange of Great Britain, Limited, the shipper and consignee, to recover loss due to decay of pears shipped in good order and condition upon appellee’s steamship New York, carried on the voyage from New York to-Southhampton. The pears were practically a total loss, and this was due to the failure-of the refrigerating plant to keep proper temperature in the storeroom. The cause of the-high temperature was due to the breaking of the port compressor rod neck bush, and this was caused by the excess of moisture in? the refrigerating machine. Appellee asserts that the excess moisture in the gas circuit of the machine froze and stopped the gas flow, and this stoppage caused the machine-to labor, causing the compressor rod to-heat up, break the bushing, burn the leathers, and incapacitate the port compressor. The *27 ■break was due to water in the gas circuit. The charge of negligence lay in introducing water into this circuit.

It was a carbon dioxide refrigerating system, using brine as a medium of transfer. The brine circuit is separate from the gas circuit. In the evaporator, the brine is •chilled by the vaporizing or boiling of the gas, and the brine is pumped through its •circuit from the evaporating chamber to the refrigeration box, and back again to the evaporator. The gas circuit has a compressor in a cylinder chamber in which moves a piston. This piston is moved by a compressor rod. Where the compressor enters the compression chamber, there is a closely fitted washer, which is called the compressor rod neck bush. The 'compressor rod is incased, and in it there are cupped leathers, between which leathers and compression chamber is the bush neck. The compressor rod easing has in it an oil pump, designed to keep an oil feed at a greater pressure than the maximum pressure in the compression chamber, ■so as to prevent any escape of gas through the ■opening between the compressor rod and the compressor rod neck bush. When the appropriate pressure has been reached, it opens an outlet valve and the gas is forced into a tube, which runs down into the separator, and from there the gas is forced through another tube, from the top of the separator to the top of the condenser, thus leaving all the oil and moisture and other elements, which are heavier than the gas, at the bottom of the separator. The condensor is the container through which run the gas tubes, and outside ■of the tubes, but within the container, is salt water.

As the gas is forced down through the condenser tubes, the exeess temperature, which is taken on by compression, is carried off by the sea water, and about the time the gas has reached the bottom of the condenser tubes it is in liquid form. It is carried through the compressor to the expansion valve by a pipe; the expansion valve is used to control the amount of gas which goes into the evaporator. It may be shut off entirely, so that no gas enters the evaporator. There is a charging cock, where fresh gas in liquid form is introduced into the system, when necessary, to compensate for leakage. Erom the expansion valve, the liquid gas is sprayed into expansion coils at the bottom of the evaporator, which is a container similar to the condenser. Conducting tubes go into and up through the evaporator and into the compressor through the intake valve. These tubes in the evaporator are called expansion coils. In the evaporator surrounding these tubes is the brine. The gas is supplied in cylindrical steel flasks in liquid form under pressure of 1,000 pounds. The flasks have only one valve, which is at the head. The contents of the flasks are never chemically pure carbon dioxide, for there is a small percentage of air, water, and other foreign substances. The percentage of water is conceded by stipulation to be 3% ounces of .water to 49 pounds.

There is expert testimony that in custom and practice the water is kept from the gas circuit by draining it from the flasks before using. The flasks are stored head down, so that the water, which is heavier than gas, will settle at the valve end, and when it becomes neeessary to increase the quantity of gas in the machine, the flasks should be blown; that is, the valve in the head should be slightly opened, and kept open until the water is blown out by the force of the compressed carbon dioxide. This is a recognized method of eliminating water from the gas as it is received on shipboard. There is also testimony that it is common practice to rig up a drier, in which the gas is run through calcium chloride, which chemical has a strong affinity for water. If the water is not thus removed by either of the methods suggested before the flasks are used, it will be free to pass into the gas circuit. While it is not altogether necessary that all of the moisture be removed, still, if these methods are pursued, a very small amount would be present, which would not impair the operation or effectiveness of the machine. Any moisture in the gas which passes the control valve is immediately subjected to a temperature below freezing, and consequently would condense and freeze in the tubes at the lower part of the evaporator, and when sufficient water or moisture has been admitted to the tubes to form a solid cake of ice sufficient to-withstand the pressure of 1,000 pounds, the circulation of the carbon dioxide gas is stopped.

This stoppage is usually gradual, and while the ice is accumulating, and the flow of gas is being restricted by the formation of ice, the chilling of the brine is also arrested. At the same time, the pressure in the lower parts of the circuit is necessarily upward. The pressure in the circuit immediately before the compressor will diminish while the pressure in the compressor will naturally increase before the stoppage. This dangerous condition is disclosed by the pressure gauges of the machine, whose fluctuations give ample warning of trouble in the system. And further evidence that the machine is out of order, *28 and that the evaporizing gas is chilling all the way up through the evaporator is disclosed by the frost dropping off the pipes which run from the top of the evaporator to the intake valve of the condenser. One effective method of removing the ice formation in the evaporator is to introduce alcohol or essence of peppermint into the evaporator coils, which melts the ice and prevents further freezing until the system can be cleared of water. Another method is to close the brine circuit, pump the brine out of the evaporator, close the control valve, warm the evaporator coils, pump the gas out of the evaporator coils, and drain out the water.

The nature of the break here was stipulated. The appellant explains how the rod neck bush was broken by saying that there was sufficient water in the gas system to freeze solid in the evaporator coil, which stopped the flow of gas and caused the compressor to operate under such greater strain that the compressor rod heated and caused sufficient friction with the neck bush and the leathers behind it to bum up the leathers, break the neck bush, and permit small particles of the carbon from the burned leather to be admitted into the compression chamber, which would score the chamber and the valve seating.

At the trial, it was stipulated that “said damage was caused by decay, which resulted solely by reason of too high temperatures in the refrigerating box, due in turn solely to breakage and failure of the refrigerating machinery. Said breakage and failure was caused by the fact that the carbon dioxide gas used in the operation of the refrigerating plant was found when used on the voyage to be defective in that it contained excessive moisture. Said gas had been supplied to the vessel by respondent purchasing it from reputable dealers of that commodity, and when said cargo was loaded and when the voyage began; respondent had exercised due diligence to make the vessel seaworthy in. respect of her supply of carbon dioxide gas. The contents of the containers were 49 lbs. of which 3% oz. (plus what adhered to the inside of the container in the laboratory test made at the end of the voyage) was water. The carbon dioxide gas was purchased in the usual containers and for the use to which it was put.”

If the carbon dioxide gas containing the named percentage of moisture was used as it came from the manufacturer, and no precaution, as is customary in the trade, was taken to drain off the moisture, it was sufficient to establish negligence of the appellee. Again by stipulation it was permissible for the appellant “to offer expert testimony, based on the facts stipulated herein, for the purpose of proving, if he can, that it was negligence and want of care, skill, and diligence to put into said refrigerating machinery and to use therein the aforesaid carbon dioxide gas containing said excess moisture, and that said excess moisture should not have been permitted to cause said breakdown.”

The appellant gave such expert testimony as to the custom and practice followed. It will be noted that the stipulation is that the gas used in the machine was found, when used on the voyage, to contain the excess water which came in the flasks. If the flasks had been inverted and the water drained off before the gas was put into' the refrigerating machine, the claim is there would have been no freezing and consequent damage. That the appellee clearly understood the stipulation to be that the gas had been introduced containing the water is significant,, in the statement of the stipulation that the mixture contained slightly in excess of 3% ounces of water per flask. This quantity of moisture was sufficient to cause the breakage, and it is stipulated that the excess temperature was due solely to the breakage and the failure of the machine. Under these circumstances, it' is no excuse to say that the gas was purchased from a reputable manufacturer. Since the presence of water in gas is a competent producing cause for the break, and if it be shown that no effort was made to draw it off, or that no inspection was made of the refrigerating machine, so as to have detected the slowly coming on freezing of the moisture in the refrigerating machine, the appellee would be guilty of negligence.

Section 1 of the Harter Act (46 USCA § 190) then becomes applicable, and the defense of the exceptions of the bill of lading as to “any latent defect in hull, machinery, or appurtenances, or unseaworthiness of the steamer, * * * provided the owners have exercised due diligence,” becomes inapplicable. The proof affirmatively shows negligence in the exercise of due care of the cargo. It was negligent to fail to drain the water, and it was also negligent to fail to observe the warning and cure the stoppage. To permit the neck bush to be broken, and the compressor thereby disabled, was the result of neglect and inattention.

The distinction between negligence as to the cargo and negligence as to the management of the ship was pointed out in Andean Trading Co. v. Pacific Steam Nav. Co. (C. C. A.) 263 F. 559, by Judge Ward, who said:

*29 “The act of opening and closing the hatches related, not to the ship, nor to the ship and cargo, though primarily to the ship. It related primarily, and indeed exclusively, to the cargo. Therefore it was not an act of management of the vessel. The Germanic [C. C. A.] 124 F. 1, 59 C. C. A. 521; Id., 196 U. S. 589, 25 S. Ct. 317, 49 L. Ed. 610. It was not an act, to use the language of the court in Knott v. Botany Mills, 179 U. S. 69, 74, 21 S. Ct. 30, 32 (45 L. Ed. 90), ‘primarily connected with the navigation or the management of the vessel, and not with the cargo.’ The District Judge disapproved the decision in The Jean Bart (D. C.) 197 F. 1002, which we think correctly construes the Harter Act.”

Negligence relating to ventilating perishable cargo is a fault relating to the cargo alone (The Skipton Castle (C. C. A.) 243 F. 523), as is the failure to properly place a tarpaulin over the hatch cover (The Edith (C. C. A.) 10 F.(2d) 684), or keeping cargoes sufficiently refrigerated (The Samland (D. C.) 7 F. (2d) 155).

The learned district judge, we think, misconstrued the terms of the stipulation. By expert testimony as to custom and practice and effect of moisture in refrigerating machines, the appellant established a case which called forth from the appellee proof as to care and inspection of the refrigerating plant, as well as the practice it followed in draining the moisture from the flasks.

We reverse the decree, with permission to the appellee, on a new trial, to be had in the District Court, to offer such proof as it may be advised to establish diligence by it in the matter of draining the flasks, and inspection and care of the refrigerating machine during the voyage.

Decree reversed.