Comparing Dehulled Hemp Meal and Canola Meal as a Protein Supplement for Lactating Dairy Cows

Effects of replacing canola meal with dehulled hemp meal in the diet of lactating dairy cows on the dry matter intake (DMI), milk production, milk fatty acid profile, blood metabolites, total-tract nutrient digest-ibilities, and transfer of cannabinoids were determined in 12 lactating, non-pregnant Holstein cows. These cows were used in a 3 × 3 Latin Square design with 3, 3-wk experimental periods, consisting of 2 wk of adaptation and 1 wk of sampling. Cows received basal partial mixed rations supplemented with either 15% DM canola meal (CM15), 15% DM dehulled hemp meal (HM15), or 7.5% DM dehulled hemp meal and 7.5% DM canola meal (CM7.5HM7.5). Diets were formulated to be isoenergetic and isonitrogenous, but the HM15 and CM7.5HM7.5 diets contained, on average 1.2%-units more CP that the CM15 diet. The CP of the dehulled hemp meal contained less soluble protein than that of canola meal. Hence, the intake of soluble protein did not differ among diets. Canola meal contained less crude fat than hemp seed meal (3.46 vs. 8.25% DM). The lipid fraction of canola meal fat contained more oleic acid (C18: 1c9; 47.3 vs, 14.9 g/100 g of fatty acids (FA)), and vaccenic acid (18:1c11; 13.7 vs. 1.2 g/100 g of FA) and less linoleic acid (C18:2 n6; 21.9 vs. 55.7g/100 g of FA), and α linolenic acid (C18:3 n3; 3.2vs. 8.9 g/100 g of FA) than the lipid fraction of hemp seed meal. The hemp seed meal contained 4.9 µg/g cannabidiol, 5.1 µg/g cannabidiolic acid, and 0.1 µg/g tertahydroxycannabinolic acid-A. Treatments did not differ in DMI, yields of milk, milk protein and milk fat, total-tract NDF digestibility, and blood plasma concentrations of β hydroxy butyrate and nonesterified FA. Apparent total-tract DM digestibility was lowest in the HM treatment, whereas the digestibility of crude protein, and the concentrations of urea, in blood, urine and milk were lowest in the CM15 treatment. Canna-binoids were not detected in urine, milk, and blood plasma. Replacing canola meal with hemp seed meal increased milk fat contents of polyunsaturated fatty acids (PUFA), which were 3.42, 3.90, and 4.25 g/100 g of FA for the CM15, CM7.5HM7.5 and HM15 treatments, respectively. Especially, the milk fat contents of 18:2 n6 (1.99 vs. 1.56 g/100 g FA) and 18:3 n3 (0.31 vs. 0.43 g/100 g FA) were increased by HM feeding. Especially, the milk fat contents of 18:2 n6 (1.99 vs. 1.56 g/100 g FA) and 18:3 n3 (0.31 vs. 0.43 g/100 g FA) were increased by hemp meal feeding Our data show that hemp seed meal is a suitable and safe replacement for canola meal as a feed for lactating dairy cows, and that this replacement increases crude protein digestibility, and urea in urine, milk and blood plasma, as well the PUFA content of milk fat.


INTRODUCTION
Between 2011 and 2020 the production of industrial hemp (Cannabis sativa L.) in Canada has increased from 15,056 to 54,941 ha (Health Canada, 2022).This has increased the availability of hemp co-products, such as hemp seed meal obtained from the oil extraction of hemp seeds.Mustafa et al. (1999) reported that hemp seed meal has higher RUP and intestinally available CP contents than canola meal, but that these differences do not affect the total available CP in ruminant diets.Substituting canola meal with hemp seed meal in the diet of sheep did not affect their DMI and total-tract nutrient digestibilities (Mustafa et al., 1999).Including 14.3% of DM of hemp seed meal in the diet of dairy cows increased their milk production (Karlsson et al., 2010).Mierlita et al. (2018) observed that feeding 180 g/d of hemp seeds and 480 g/d of hemp meal increased yields of milk and milk fat of dairy ewes.These authors also found that addition of hemp to the diet increased the milk fat contents of vaccenic acid (18:1c11) and conjugated linoleic acids.Addo et al. (2023) also observed that replacing canola meal with dehulled hemp meal in the diet of nonlactating dairy cows did not affect their feed intake, apparent total-tract digestibilities of DM and CP, and the blood plasma concentrations of glucose, urea, β-hydroxybutyrate, and nonesterified fatty acids (FA).
Despite the research conducted on this feed, the use of hemp meal as a feed for ruminant livestock has not yet been approved in Canada.One reason for this is the concern about the transfer of psychoactive cannabinoids, such as cannabidiol (CBD) and d9-tetrahydrocannabinol (THC), into edible products, such as body tissues and milk (Ware and Tawfik, 2005).As a result, much of the produced hemp meal needs to be discarded, which poses an environmental issue and the loss of a potentially high-quality feed.Hemp meal also contains tetrahydrocannabinolic acid (THCA) and cannabidiolic acid CBDA) that can be converted to THC and CBD, respectively, by decarboxylation.Addo et al. (2023) showed that feeding up to 15% of DM as hemp seed meal to nonlactating dairy cows did not result in detectable concentrations of CBD and THC in blood, urine, rumen digesta, as well as adipose and muscle tissues.Also, feeding diets containing up to 18% DM hemp meal for 35 d did not result in detectible levels of THC in edible broiler tissues (Rasool, 2018).
Despite earlier research, the quality and safety of hemp seed meal as a protein supplement for lactating dairy cows have not yet been fully determined.This is needed to obtain approval from the Canadian Food Inspection Agency to feed these hemp co-products to dairy cows, and to convince the dairy industry about the quality of these feeds.The objectives of this study were, therefore, to determine the effects of replacing canola meal with dehulled hemp seed meal on the DMI, the production and composition of milk, concentrations of blood metabolites, apparent total-tract digestibilities of nutrients, and concentrations of cannabinoids in biological fluids in lactating dairy cows.

Animals, diets, and experimental procedure
Five primiparous and 7 multiparous non-pregnant, lactating Holstein cows with an average body weight of 673 ± 96 kg (mean ± SD) and days in milk of 114 ± 64 d (mean ± SD), and body condition scores of and between 2.5 -3.25 (1 −5 scale) were used.The number of replicates per treatment was based on treatment effects on DMI reported by Karlsson et al. (2010), including a treatment difference of 3.1 kg, a SD of 2.8 kg, a probability of 95%, and a power of 80%.The cows were housed within the large animal metabolism unit at the Glenlea Research Station of the University of Manitoba.Cows were kept in individual tie stalls lined with rubber mats, and chopped straw was used as bedding.All procedures on the use of these cows were pre-approved by the Fort Garry Campus Animal Care Committee of the University of Manitoba (protocol F21-003), and the cows were cared for in accordance with the Canadian Council on Animal Care guidelines (CCAC, 2009).The dehulled hemp seed meal used in this trial was obtained from Hemp Oil Canada Inc., Ste Agathe, Manitoba, Canada.
Whole hemp seeds were cold pressed to extract the oil, no solvent was used.The processing temperature reached a maximum 75°C for the 2 min duration of the pressing.This rise in temperature was due to friction of the seed moving through the press.The resulting seedcake was cooled and milled with a pin mill.Subsequently, hulls were physically sifting out to increase the protein content.The hemp was of the CRS-1 variety, and was grown in western Manitoba.The canola meal had been solvent extracted (Mejicanos et al., 2016) and was obtained from Trouw Nutrition in Otterburne, MB, Canada.
Cows were blocked according to their parity, previous milk production, and days in milk and were used in a repeated 3 × 3 Latin Square Design with 3, 3-wk experimental periods.The first 2 wk of each experimental period were used as adaptation, whereas the third week was used for sampling.Cows were randomly assigned to one of the following treatments: (1) a basal partial mixed ration (PMR) diet plus 15% DM canola meal (CM15 treatment); (2) a basal PMR diet plus 15% DM hemp seed meal (HM15 treatment); or (3) a basal PMR diet plus a 7.5% DM hemp seed meal and 7.5% DM canola meal (CM7.5HM7.5 treatment).Diets were formulated to be isoenergetic and isonitrogenous.Cows were fed equal amounts twice daily at 0930 h and 1600 h, allowing for between 5 and 10% of feed refusals.The hemp and canola meal were weighed out for individual cows for each feeding based upon the amount of the base ration delivered to each cow, so that the desired dietary inclusion rates of the meals were maintained.The meals were added to the base diet in each cow's feed bunk, and thoroughly mixed with the base diets.This mixing prevented cross-contamination of meal in the feed mixer.
Cows had free access to fresh water.The ingredient composition of experimental diets and protein supplements are presented in Table 1.The chemical composition of forages used in the experiment are presented in Table 2.The chemical composition of experimental diets and the protein supplements are given in Table 3.Then FA profile of diets and protein supplements are given in Table 4. Cows were milked twice a day at approximately 0600 h and 1700 h.
Dry matter intake, body weight, and feed analyses.Samples of diets were collected once daily at the morning feeding during each experimental period, pooled by treatment and stored at -20°C.Individual orts samples were collected every morning immediately before feed delivery, stored in a -20°C, and pooled by weight for individual cows after the last week of each experimental period.
Pooled diet and orts samples were dried in a forced-air oven at 60°C for 48 h.The dried samples were ground using a Wiley mill through a 1-mm screen (Thomas-Wiley, Philadelphia, PA) and were stored at room temperature until further analyses.The DMI was determined using the amounts and DM contents of PMR and protein supplements offered, and orts.Samples were analyzed for DM and CP (AOAC 990.03;AOAC 2019).Soluble protein (SP) was analyzed according to Licitera et al. (1996) and the protein that dissolved in a borate-phosphate buffer.Neutral detergent fiber (NDF) was analyzed according to Van Soest et al. (1991) using α-amylase (Sigma No. A3306; Sigma Chemical Co., St. Louis, MO, USA) and sodium sulfite, and corrected for ash concentration, but not for protein, using an Ankom 200 Fiber Analyzer (Ankom Technology, Fairport, NY, USA).Acid detergent fiber (ADF) was analyzed using AOAC method 973.18 (AOAC 2019) and the Ankom 200 Fiber Analyzer (Ankom Technology, Fairport, NY, USA).Starch was measured using a UV method (method 996.11;AOAC 2019).Acid insoluble ash (AIA) was analyzed using AOAC method 920.08 (AOAC 2019).Ether extract and ash analyses were conducted using AOAC methods AOAC 920.39 (AOAC 2019) and 923.03 (AOAC 2019), respectively.Inductively coupled plasma emission spectroscopy (AOAC method 968.08;AOAC 2019) using a Plasma Spectrometer (Thermo Jarrell Ash Corp., Grand Junction, CO, USA) and acid digestion were used to analyze calcium, phosphorous, potassium, magnesium, and sodium.
Determination of FA composition of experimental diets, canola meal, and hemp seed meal was carried out by gas chromatography (GC) according to the method described by Villeneuve et al. (2013) and Fauteux et al. (2016).One-step transesterification of fatty acids in milk was conducted by incubation with sodium methoxide (0.5 M in methanol) at 70°C for 60 min, followed by another incubation with HCL (10% in methanol) at 50°C for 30 min.A Clarus 680 gas chromatograph (Perkin Elmer, Waltham, MA), a HP-Innowax polar capillary column (30-m length, 0.320 mm i.d., 0.25 µm film thickness; Agilent Technologies Canada Inc., Mississauga, ON, Canada), and a flame-ionization detector were used for the detection of fatty acid methyl esters.The column temperature at sample injection was 185°C, and was maintained 0.5 min.Subsequently, this temperature was increased at 3°C/min to 220°C.The temperatures at the Inlet and detector were at 240 and 250°C, respectively.The split ratio was 50:1m and the hydrogen carrier gas flow rate was 1 mL/min.The DMI, and the CP, NDF and AIA contents of experimental diets and pooled feces of individual cows during the last week of each period were used for the estimation of the apparent total-tract digestibility coefficients (ADC) of nutrients.The AIA content of diets was used as an internal digestion marker to determine the ADC of DM, CP, and NDF (Van Keulen and Young, 1977;McGeough et al., 2010).Body weight of cows were determined on 2 d at the beginning and the end of the experiment, and at the beginning and end of each experimental period.

Milk yields and milk analyses.
During sampling weeks milk yields were determined using Tru-Test regulation meters (Westfalia Surge, Mississauga, ON).Milk samples were obtained directly from the regulation meters.Two mL of milk were collected into a pre-labeled 2 mL tube at each milking for 4 consecutive milkings for each cow, and immediately flash frozen in liquid nitrogen and later sustained at a temperature of −80°C until further analysis for cannabinoids.For milk composition analysis, 40 mL of milk were collected from the regulation meter into a 50 mL vial from each cow and stored in a fridge until all 4 consecutive milkings have been completed.Samples were later warmed and vortexed and aliquots were pooled in 50 mL vials in a yield-toweight ratio and stored at 4°C until further milk com- ponents analyses at Horizon Laboratories (Winnipeg, MB, Canada).These samples were analyzed for fat, crude protein, lactose, milk urea nitrogen (MUN) and β hydroxybutyrate (BHBA) by mid-infrared analysis using a Milk-o-Scan 303AB (Foss Electric, Hillerød, Denmark).Milk fat extraction was carried as described by Boivin et al. (2013) and Rico et al., (2021) by adding a 25% ammonium solution, ethanol, diethyl ether, and petroleum ether to the milk sample during intermittent vortex mixing.Following the formation of 2 distinct phases, the upper phase containing the extracted fat was removed for methylation.The methylation was conducted by adding hexane, methyl acetate the methylation reagent (methanol and sodium methoxide solution), termination reagent (oxalic acid and diethyl ether) and calcium chloride to the milk fat extract during intermittent vortex mixing.Subsequently, the newly formed methylated extract was removed for gas chromatography (GC) analysis.
Milk FA composition was determined according to the procedure described by Boivin et al. (2012) by GC (Agilent 7890A; Agilent Technologies) using a 100-m CP-Sil-88 capillary column (0.25 mm i.d., 0.20 mm film thickness; Agilent Technologies Canada Inc.) and a flame ionization detector.At the time of sample injection, the column temperature was 80°C for 1 min, then increased at 28°C/minute to 215°C, and maintained for 21.5 min.When isomers coeluted under the first temperature program, then isothermal temperatures of 150°C and 175°C were used.
Fecal, urine, and blood sample collection and analyses.Fecal grab samples were collected twice a day at 0900 h and 1500 h over a 5-d period during the last week of each experimental period.Fecal sample collection was conducted as described by Li et al. (2012).Approximately 250 g of fecal grab samples were collected from the rectum of each cow and stored at −20°C.Samples were later thawed and oven-dried at 60°C for During the third week of each experimental period on Days 15 and 17 of experimental periods urine and blood samples were collected twice daily at 0900 h and 1500 h.Approximately 80 mL of mid-stream urine samples were collected by stimulating the perirenal area to initiate urination.After collection, 30 mL of urine sample was transferred into tubes pre-filled with 2 mL of 3N hydrochloric acid (HCl), to ensure that the pH of the urine samples was below 2.0 and to minimize N volatilization.The samples were later stored in a −20°C freezer until final analysis for urine nitrogen using procedure AOAC 990.03 (AOAC 2019).Approximately 2 mL of urine from the initial sample was transferred in 2 mL screw-top vials, immediately flash frozen in liquid nitrogen, and later stored at −80°C until analysis for cannabinoids.
Blood samples were taken from the tail vein and collected into 10 mL vacutainer tubes for serum and 10 mL heparinized vacutainer tubes for plasma analyses.
For blood serum extraction, collected blood in vacutainer tubes was kept at room temperature for 30 min to ensure clotting before being centrifuged at 1900 g at 4°C for 10 min.Serum samples were later transferred into plastic tubes and stored immediately in a −20°C freezer until further analyses.The analysis of serum samples was conducted by Veterinary Diagnostic Services (Manitoba Agriculture, Food and Rural Initiatives, Winnipeg, MB).Samples were analyzed for nonesterified fatty acids (NEFA) and BHB using the RX Monza analyzer and Randox kits (Randox Laboratories Ltd., Crumlin, UK).Serum glucose and urea were analyzed using the Cobas c 502 analyzer and kits for this instrument (Roche Diagnostics GmbH, Mannheim, Germany).Blood samples collected for plasma analysis were placed on ice and centrifuged at 4°C for 15 min immediately after collection.Plasma samples were then transferred into prelabelled 2mL tubes and immediately flash frozen in liquid nitrogen and later kept at a temperature of −80°C until analysis for cannabinoids.
Cannabinoid analyses.Sample preparation and quantification of cannabinoids in feed, feces, blood plasma, urine and milk samples were conducted by In-notechAlberta in Vegreville, AB, which holds a license for analytical testing of cannabinoids under the Canadian Cannabis Act (Government of Canada, 2018), as described by Addo et al. (2023).Dried and ground feed and feces samples were mixed with HPLC grade methanol (1 g/ 5 mL), sonicated for 12 min (feces samples only), shaken (1500 stroke/min), and centrifuged for 12 min at 250 g.Blood plasma samples were mixed with HPLC grade methanol (100 µL/200 µL) and vortexed for 10 min at 1600 g.Urine samples were mixed with water (200 µL/400 µL) and vortexed.Subsequently, 160 µL of 10% acetic acid were added, and the mixture was vortexed again.This was followed by adding 1.6 mL of a 9:1 hexane: ethylacetate solution and sonicated for 30 min, and centrifugation at 280 g for 15 min to achieve separation.Subsequently, the organic upper layer was transferred, the fluid was evaporated, the samples were reconstituted in 200 µL of a 50:50 acetonitrile: water solution, and the reconstituted samples were centrifuged for 10 min at 1600 g.Milk was mixed with cold acetonitrile (200 µL/ 400 µL), vortexed for 1 min, centrifuged for 20 min at 1600 g, and filtered through a 0.22 µm filter.
Standards of CBD (C-045), CBDA (C-144), THC (T-005) and THCA (T-093) were obtained from Sigma (Sigma-Aldrich, St. Louis, MO).These were serially diluted to provide standard solutions of 1, 2, 5, 25, 100, 200, 1000 ng/mL.Calibration curves were linear in the range from 1 to 1000 ng/mL for all cannabinoids with a coefficient of correlation greater than 0.99.The limit of detection was 2 ng/g for feed and feces samples, and 1 ng/mL for blood plasma, urine and milk samples.Average intra-day and inter-day accuracies were 80 -120% and 70-130%, respectively.For spiking and recovery analysis, cannabis free matrices were spiked with 0.05 µg/g of each cannabinoid for feed and feces samples, whereas rumen fluid, blood plasma, milk and urine samples were spiked with 0.01 µg/g of these cannabinoids.

Statistical analysis.
Data were analyzed using a repeated Latin square model by the MIXED procedure of SAS (Ver.9.4 SAS Institute Inc., Cary, NC). using the following model: Where Y ijkl = observation of dependent variables, µ = overall mean, C i = random effect of cow, P j = random effect of period, B k = random effect of block, T lk = fixed effect of treatments (CM15, HM15, and CM7.5HM7.5)and e ijkl = residuals.
The effect of treatment was considered fixed.The effects of square, cow within square, and period were considered random.For dependent variables that had repeated measurements, the repeated measurement option within the SAS MIXED procedure (Ver.9.4 SAS Institute Inc., Cary, NC) was used.In case of unequal variances, the (UN), (CSH) and (CS) covariance structures for DMI and concentrations of blood metabolites, were included in the REPEATED statement to allow for separate estimation of variances.If the assumption of equal variances was not met, the DDFM = KR option of the PROC MIXED procedure was added to the model to adjust the degrees of freedom of the unequal variances.The PDIFF statement in SAS was used to determine differences between treatments.Comparison of least squares means was conducted with the Tukey test.Significant effects of treatments as well as interactions were discussed at P < 0.05, and tendencies were reported at 0.05 < P < 0.10.

RESULTS
Diets were formulated to be isoenergetic and isonitrogenous (Table 2).The hemp seed meal contained more CP (56.7 vs. 43.6%DM), and crude fat (8.25 vs. 3.45%DM) and less NDF (23.7 vs. 29.0%DM)than canola meal.Hemp seed meal protein contained less soluble crude protein than canola meal protein (19.2 vs. 25.5% CP).However, due to the higher CP content of the hemp seed meal, the CP content of this meal was only 0.23 percentage points lower that than of canola meal.Although the diets were formulated to be isoenergetic and isonitrogenous, the HM15 and CM7.5HM7.5 diets contained, on average 1.2%-units more CP than the CM15 diet.
Replacing canola meal with hemp seed meal did not affect milk yields or the milk contents of fat, protein, lactose and BHBA (Table 4).However, replacing canola meal with hemp seed meal increased the MUN content of milk from 14.3 to 15.6 mg/dL (P < 0.05).This replacement did not affect DMI, and the total-tract NDF digestibility (Table 5).However, replacing canola meal with hemp seed meal reduced the apparent DM digestibility from 68.5 to 66.1% (P < 0.05) and increased the apparent CP digestibility from 64.6 to 67.4% (P < 0.05).For all treatments, the urine, milk, and blood plasma did not contain any detectable concentrations of the cannabinoids CBD, CBDA, THC and THCA, which implied that concentrations of CBD and CBDA were below 0.048 ng/mL, and below 0.024 ng/mL for THC and THCA.Treatment did not affect the blood plasma concentrations of BHBA and NEFA, but the HM15 treatment had higher blood urea 5.95 vs. 5.31 mmol/L (P < 0.05), blood glucose 3.65 vs 3.56 mmol/L (P < 0.05), and urinary N contents (0.90 vs. 0.77%, P < 0.05) than the CM15 treatment.(Table 7).

Feed quality and diet composition
The objective of the present study was to compare the effect of replacing canola meal with dehulled hemp seed meal on milk production and nutrient utilization of dairy cows and the transfer of cannabinoids of these cows.The main differences between the canola meal and the dehulled hemp seed meal used in this study were that the hemp seed meal contained more CP (56.7 vs 43.6%DM) and crude fat (8.3 vs. 3.5%DM) and less NDF (23.7 vs, 29.0%DM) and SP (19.2 vs. 25.5%CP)et al., 2021).The hemp seed meal used in our study contained a higher amount of CP than these industry averages.This was likely due to the sifting out of hulls.
The difference in SP agrees with Mustafa et al. (1999), who reported that CP of hemp meal has a lower rumen degradability than that of canola meal.Bell and Keith (1991) reported averages of the CP content of canola meal from several commercial plants of 41.9% of DM.Hence, the CP content of the canola meal used in our study is comparable to these averages (Table 10).
The fatty acid profile differed between the 2 meals, in that the canola meal contained 1.59 times less linoleic acid (C18:2 n6) and 2.17 times more oleic acid (C18:1 cis-9), than hemp seed meal.However, due to the higher crude fat content of the hemp meal, the cows on the CM15 treatment only ingested 1.17 times more oleic acid (18:1 cis-9), compared with the cows on the HM15 treatment.The cows on the HM15 diet, however, ingested 4.73 times more linoleic acid (C18:2 n6) than cows on the CM15 treatment.To make the diets isoenergetic and isonitrogenous, the composition of the PMR differed among treatments.This was done to prevent differences in animal performance among treatments that were due to differences in the energy and protein contents among these diets.Despite this, the HM15 and CM7.5HM7.5 diets contained, on average 1.2%-units more CP than the CM15 diet.

Feed intake
Feeding hemp seed meal instead of canola meal to lactating dairy cows did not affect their DMI.This agrees with Mustafa et al. (1999), who observed that replacing 20% of DM of canola meal with hemp meal in the diets of growing lambs did not affect DMI.Also, Abrahamsen et al. (2021) found that including up to 33% DM of hemp seed meal in the diet of growing meat goats did not affect their DMI.However, as adding hemp seed meal increased the NDF, ADF and lignin contents, and decreased the starch and NEl content of the diet, this addition reduced the growth performance of these goats.Hessle et al. (2008) replaced 1 kg of hemp meal with 0.5 kg of barley grain and 0.5 kg of  soybean meal in the diet of calves, and replaced 0.2 kg of hemp meal with 0.1 kg of barley grain and 0.5 kg of soybean meal in the diet of finishing steers.This replacement increased the DMI of calves, but the DMI of steers was not affected.These replacements did not affect live weight gain of the calves and steers.As a result, hemp meal feeding reduced the feed efficiency in calves.Gibb et al. (2005) included up to 14% of full-fat hemp seed in the diet of feedlot cattle, and observed that this inclusion did not affect DMI, average daily gain, and the gain to feed ratio.Also, Karlsson et al. (2010) replaced up to 32% DM of a basal diet with hempseed meal, and found that this replacement did not affect DMI of lactating dairy cows.However, due to differences in the composition of the basal diet and the hemp seed meal, this replacement increased the intakes of CP, crude fat and NDF, and decreased that of starch.A comparison of these studies showed that differences in inclusion levels and composition of basal diets affect the impact of the addition of hemp meal on the DMI.However, these studies agree that hemp meal is palatable, and that when this meal is included up to 15% DM in the diet of ruminants, this does not affect feed intake and growth.

Milk production
In the present study replacing canola meal with hemp seed meal did not affect daily milk yields, as well as the contents and amounts of milk fat and milk protein.This agrees with earlier studies on feeding hemp meal instead of soybean meal and concentrate pellets to small ruminants like sheep and goats (Mierliţă, 2016;Mierliţă, 2018;Šalavardić et al., 2021).Karlson et al. (2010) fed either 0, 143, 233, or 318 g/kg DM of hemp meal to lactating dairy cows and found that feeding this meal increased the daily yields of milk, milk fat and milk protein.They obtained the highest yields of milk, milk fat and milk protein with a hemp meal inclusion of 143 g/kg DM of the diet.This finding contrasts with our study, but in the study from Karlson et al. (2010) feeding hemp meal increased intakes of CP, crude fat, and NDF, while decreasing that of starch.Hence, in the earlier study, the changes in milk production may  9. Effects of experimental diets (CM15 = 15% DM canola meal; CM7.5HM 7.5 = 7.5% dehulled hemp seed meal and 7.5% DM canola meal; HM15 = 15% DM hemp seed meal) on milk fat proportions of saturated and mono-unsaturated long-chain fatty acid proportions of lactating dairy cows (g/100 g of FA) x, y Least squares means with different superscripts within rows tend to differ (P < 0.10).
1 SEM = standard error of mean.
have been due to the altered intakes of these nutrients, rather than due the increased intake of hemp meal.
Replacing canola meal with hemp seed meal increased the concentration of MUN from 14.3 mg/dl to 15.5 mg/ dl (P < 0.05).This was likely due to the increase in the blood urea nitrogen content, which was increased from 5.31 to 5.95 mmol/L by this replacement.These increases were likely the result of higher rumen ammonia nitrogen concentrations.Although, diets were formulated to be isoenergetic and isonitrogenous, the HM15 and HM7.5CM7.5 contained, on average 1.2%-units more CP than the CM15 diet.Hence.dietary differences in MUN may have been due to dietary differences in the CP content.However, as the SP content of CP was lower in hemp meal than in canola meal, the intake of SP did not differ among diets.Hence, dietary differences in the composition of CP may also have contributed to the dietary differences in MUN.The latter was suggested by Karlsson et al. (2010) and Šalavardić et al. (2021) who reported that the MUN concentration increased and that the efficiency of converting dietary CP into milk protein decreased when the inclusion rate of hemp meal in the diets of lactating cows and lactating goats increased.The higher starch content of the CM15 diet compared with the HM15 diet could have affected microbial protein synthesis in the rumen and the utilization of rumen ammonia nitrogen.However, as we did not determine this synthesis, this could not be proven.Jonker et al. (1998) recommended to target MUN concentration between 10 to 16 mg/dl, depending on the milk production yield.Kohn et al. (2002) revised these targets to be between 8.5 to 11.5 mg/dl.Hence, for both canola meal-and hemp seed meal-fed cows, the MUN concentrations were toward the upper end of recommendations.
In our study, the diets were only fed for 2 wk before effects on the yield and composition of milk were determined.A longer adaptation to the diets could have altered the impact of the treatments on the yield and composition of milk.We decided for a 2 wk adaptation, as long adaptation and experimental periods increase the risk of treatment by period interactions.However, after 2 wk of feeding, effects on the milk fatty acid pro- file and MUN could already be observed.Also, Wagner et al. ( 2022) observed that feeding a high cannabinoid diet for 7 d to lactating cows already resulted in detectable concentrations of cannabinoids in milk.Hence, we believe that in our experiment, effects of experimental diets on the yield and composition of milk could already be detected after 2 wk of feeding these diets.

Total-tract nutrient digestibilities
In our study, replacing canola meal with dehulled hemp seed meal reduced the apparent total-tract digestibility of DM, increased that digestibility of CP, while not affecting the digestibility of NDF.This contrasts with an earlier study by Addo et al., (2023) that showed that this replacement did not affect the apparent totaltract digestibility of DM and CP, but reduced that of NDF.Another difference between these studies is that the diets of the earlier study contained less CP, but more NDF.Mustafa et al. (1999) and Semwogerere et al. (2020) also observed that the apparent DM and CP total-tract digestibilities of hemp seed meal and canola meal did not differ.Several studies have, however, reported that the rumen degradation of hemp seed meal is lower and slower than that of hemp meal (Mustafa et al., 1999;Krizsan and Huhtanen, 2013;Semwogerere et al., 2020), However, this slower rumen NDF degradation rate may be compensated by an increase in rumen retention time (Semwogerere et al., 2020).A difference between the studies from Mustafa et al. (1999), Krizsan and Huhtanen (2013), and Semwogerere et al., 2020 and our study is that in our study hulls were sifted out of the original hemp seed meal, thereby increasing the CP content and reducing the NDF content of the remaining meal.This sifting may also have altered the rumen degradability and digestibility of the CP and NDF (Ishler and Varga, 2008).Hence, the differences between the total-tract digestibilities of our study compared with those of earlier studies may be related to the relatively high CP and the relatively low NDF contents of the diets and hemp seed meal in our study.

Milk fatty acid profile
In agreement with earlier studies, the crude fat fraction of hemp seed meal fatty acids contained a higher proportion of PUFA and a lower proportion of MUFA compared with canola meal (Teh and Birch, 2013;Mungure and Birch, 2014).Especially, the concentrations of 18:2 n6 and 18:3 n3 were higher and those of 18:1 c9 and 18:1c11 were lower in dehulled hemp seed meal than in canola meal.These findings agree with the studies of Birch (2013), Mungure and Birch (2014) and Mierliţă (2018).The differences in the lipid composition between the meal explain why hemp seed meal-fed cows had higher milk fat proportions of 18:2 n6 (1.99 vs. 1.56 g/100 g FA) and 18:3 n3 (0.43 vs. 0.31 g/100 g FA) than canola meal-fed cows.These higher concentrations in PUFA could also explain why feeding hemp seed meal increased milk fatty acid proportions of rumen biohydrogenation intermediates.Despite the higher proportion of MUFA in the lipid fraction of canola meal compared with hemp seed meal, canola meal-fed cows did not have higher milk fat proportions of these fatty acids than hemp seed meal-fed cows.This is explained by the lower crude fat content of canola meal compared with hemp seed meal (3.46 vs. 8.25%DM).Hence, the intake of MUFA did not differ substantially among the treatments.

Blood metabolites
Both blood plasma concentrations of NEFA and BHBA are used as markers of a negative energy balance (McArt et al., 2012).Also, the blood plasma concentration of glucose can be used as such a marker, and glucose is needed to provide energy to cells and tissues of the cow, and for the production of lactose (Aschenbach et al., 2010).Despite this, gluconeogenesis and the cellular uptake of glucose are under hormonal control (Aschenbach et al., 2010), which may limit the usefulness of blood glucose as index of energy status.Treatments did not differ in the blood plasma concentrations of NEFA and BHBA, which suggested that the replacement of canola meal with hemp seed meal did not affect the energy status of the cows.This agrees with Šalavardić et al. (2021) who observed that replacing 60 or 120 g of soybean meal and extruded soybeans in a diet of lactating goats with hemp meal did not affect their blood plasma concentrations of NEFA and BHBA.The cows of our study were all beyond early lactation.Hence, high blood plasma concentrations of NEFA and BHBA and treatment differences in these concentrations were not expected.
The cows on the HM15 treatment had higher blood glucose concentrations than cows on the CM15 treatment (3.65 vs. 3.56 mmol/L).As diets were formulated to be isoenergetic, treatment differences in the intake of NEl were unlikely the reason for this treatment difference.The dietary NFC and starch contents of the hemp seed meal-fed cows were lower than those of canola meal fed-cows, and the DMI did not differ among diets.Hence, dietary differences in the intakes of NFC and starch cannot explain these differences in blood glucose concentrations either.It is possible that hemp seed meal-fed cows had a higher availability of glucogenic amino acids.However, as we did not measure these amino acids, this cannot be proven.

Cannabinoids
One of the concerns of feeding hemp by-products, such as hemp seed meal, to ruminants is the transfer of cannabinoids to edible products, such as meat and milk.The legal THC limit in commercial hemp products in Canada is 10 ppm (Chicoine et al., 2020;Meng et al., 2018).The THC content of the hemp seed meal used in our study was well below that limit.In our study, 69.4 and 63.4% of the ingested CBD and CBDA, respectively, were excreted by for cows on the HM7.5CM7.5 diet.For the cows on HM15 diets, 63.5 and 52.5% of the ingested CBD and CBDA, respectively, were excreted in the feces.In our earlier study with nonlactating cows (Addo et al., 2023), 102.2 and 68.7%, respectively, of the ingested CBD and CBDA were excreted in the feces.Hence, in both studies the majority of ingested CBD and CBDA were excreted in the feces.Kleinhenz et al., (2020) demonstrated that a onetime oral administration of industrial hemp flower material to cattle with a CBDA dose of 5.4 mg/kg of BW resulted in a maximum blood plasma concentration of CBDA of 72.9 ng/ml at 11.8 h after dosing, and a maximum THCA blood plasma concentration of 12.1 ng/mL at 25.2 h after dosing.Also, intakes of 1.6, 0.09, 10.7 mg/kg BW of THC, THCA and CBD by feeding hemp silage to lactating dairy cows resulted in blood plasma concentrations of 10.7, 12.9, and 20.7 ug/kg, respectively (Wagner et al., 2022).When these intakes were increased to 1.6, 0.09, 10.7 mg/kg BW of THC, THCA and CBD, respectively, the blood plasma concentrations were increased to 12.5, 13.1, and 35.1 µg/kg, respectively.This shows that cannabinoids can be transferred from hemp feeds to blood of dairy cows.However, the intakes of THC, THCA and CBD in our study did not exceed 0, 0.0005, 0.023, and 0.027 mg/kg BW of THC, THCA, CBD, and CBDA, respectively.Hence, the intakes of THCA and CBD in the study from Wagner et al. (2022) were 180 and 465 times, respectively, higher than those in our study.This explains why no detectable concentrations of these cannabinoids were found in the blood plasma of cows on our trials.Wagner et al. (2022) also determined the transfer of cannabinoids from feed to milk of lactating dairy cows.During an adaptation period, cows received between 0.31 and 0.92 kg DM per cow of a hemp silage that contained 58.3, 7.4, and 804.7 mg/kg DM of THC, THCA and CBD, respectively.This did not result in detectable concentrations of these cannabinoids in milk.However, feeding 0.84 kg DM per cow of a hemp silage containing 1,254, 70.1 and 8,304 mg/kg DM of THC, THCA and CBD, respectively, increased the contents of THC, THCA and CBD in milk to 87.9, 0.51, and 278 ug/kg, respectively.When the feeding of this hemp silage was increased to 1.68 kg DM per cow, the concentrations of THC, THCA and CBD in milk were increased to 243, 0.34, and 982 ug/kg, respectively.Based on these results, the feed-to-milk transfer rates of THC, THCA and CBD were estimated to be 0.20, 0.025, and 0.043% for THC, THCA and CBD, respectively.Even during the adaptation period of the study from Wagner et al. (2022), the intake of CBD was 40.1 times greater than that of the cows on the HM15 treatment in our trial.That, combined with an estimated feed-to-milk transfer rate of CBD of 0.043%, explains why in our study none of the treatments resulted in detectable concentrations of cannabinoids in the milk.

CONCLUSIONS
The replacement of canola meal with hemp seed meal as protein supplements in diets for lactating dairy cows, did not alter their DMI, milk yield, milk protein, milk fat, total-tract NDF digestibility.The effects of this replacement on milk fatty acid composition were limited to an increase in 18:2 n6, 18:3 n3, and rumen biohydrogenation intermediates and a decrease in 18:1 c11.However, feeding hemp seed meal instead of canola meal led to increases in the concentration of milk total PUFA and serum glucose, and the total-tract CP digestibility.The concentration of cannabinoids in milk, blood plasma, and urine of the cows were below detection limits of 0.048 ng/mL and 0.024 ng/mL for CBD and THC, respectively.This shows that feeding hemp seed meal to lactating dairy cows does not lead to accumulation of cannabinoids in body tissues and biological fluids.Our data indicate that hemp seed meal is a good and safe alternative for canola meal as a protein supplement for lactating dairy cows.

Table 2 .
Chemical composition of forages used in the experiment