Livestock Research for Rural Development 15 (2) 2003

Citation of this paper

A note on the chemical composition and in vitro digestibility of common vegetables utilised in urban dairy systems of the east of Mexico City

L Arias, J Contreras, H Losada, D Grande, R Soriano, J Vieyra, J Cortés, J Rivera

Area of Sustainable Agricultural Development. Department of Biology of Reproduction.
Division of Biological and Health Sciences, Autonomous University Metropolitan-Iztapalapa.

Av. San Rafael Atlixco  No 186. Col.
Vicentina. Iztapalapa. C.P. 09340. Mexico D.F.


A study of eight vegetable wastes and sun-cured Lucerne used as forage sources in urban systems of production of milk of the east of Mexico City was carried out to analyze their chemical composition, fractions of the protein and dry matter and organic matter digestibility coefficients.


The contents of DM were low in three vegetables (5.7, 7.9 and 8.7% for nopal-legume, lettuce and radish leaf, respectively). The ranges obtained for the crude protein, crude fat and crude fiber went from 4.5 to 27.5%, 0.5 to 2.9% and 11.8 to 27.9% in the same order. The contents of ash were low in the case of the maize cobs (2.1 and 2.7%) and high for the lettuce and radish leaf (28.5 and 30.7%). The protein fractions ranged from 33 to 69% for the soluble protein and  0.4 to 6.2% for the fraction of the non protein nitrogen. The digestibility coefficients of DM and OM ranged from 61.3 to 85.2% and 66.4 to 99.2%, respectively. The OM digestibility was related positively with the crude protein content and in negative form with the fiber content. 


The low content of DM of most of these feeds is discussed in relation to possible effects upon voluntary intake of dairy cattle.   

Key words: feed composition, digestibility, dairy cattle, urban agriculture   


The Central Food depot of Mexico City located in the delegation of Iztapalapa, is the largest market in the world receiving 24, 000 tonnes of food products daily, some 40% of the Mexican national harvest. It generates 800 tons of waste per day, mostly organic, of which 100 tons are used as the main forage source to feed the 2500 dairy cattle maintained in urban stables (Grande et al 1994;  Losada et al  2000). A wide range of products is used as feed for the cows (up to 90% of the forage source), including parts or whole of the following: cabbage leaves (Cucurbita pepo), maize cob (Zea mays), cauliflower (Brassica oleracea var botrytis), maize cob ear (Zea mays), radish leaves (Raphanus sativus), lettuce (Lactuca sativa), nopal-legume (Opuntia ficus índica) and sun-cured Lucerne (Medicago sativa). In spite of the importance that this feeding source represents for the dairymen of the area, the information related  to the feed composition and digestibility is restricted. The objective of the present work was to study the chemical composition and the in vitro digestibility of the main wastes of vegetables used by milk producers of Iztapalapa.      

Materials and Methods   

Samples and their processing     

The feeds were obtained from 5 dairy stables located in Iztapalapa at the east of Mexico City.  The sampling included an additional sample from the Central Food depot in the same area constituting the main (and only) source of vegetable rejects for urban dairy systems in the area. An amount considered as a representative sample (2 kg, fresh basis per stable) was obtained. The sampled feeds included: cabbage and cauliflower leaves, maize cobs, maize cob ears, radish and lettuce leaves and nopal-legume. An extra sample included sun-cured lucerne in view of its wide use (secondary) in the diet of the animals. The vegetable samples were pooled to obtain a 12 kg (fresh basis) total sample of each one and later chopped and dried in an air-forced stove (Felisa FE144A) at 40-50° C during 48 h. Afterwards they were milled in a Wiley type mill, with a mesh of 1mm diameter and stored until analysis.    

Ruminal liquor sampling and elaboration of the digestive liquid   

Ruminal liquor was extracted from a 7 year-old dairy cow fistulated in the rúmen and fed with waste of vegetables, sun-cured Lucerne and concentrate. The ruminal liquor was transported in a thermos flask at a temperature of 38-39° C and after measurement of pH was filtered through two  layers of muslin and stored in a water bath at a temperature of 39° C until the moment of its use. The digestive liquid for the in vitro test was composed of 324 ml of ruminal liquor blended with 261 ml of a buffer solution and 1029 ml of distilled water. The buffer solution and the water were previously heated separately to a temperature of 38° C and later blended and bubbled with CO2  until the final solution turned clear.   

 In vitro digestibility   

The technique used was the one described by Tilley and Terry (1963) modified by Ruiz (1995).  Triplicate samples of 0.5 g of each feed  was placed in a centrifuge tube of polypropylene (NALGENE) of 100 ml. Three blank tubes (without sample) were used in each run. To each sample was added  1 ml of a solution of ammonium sulfate and 50 ml of digestive liquid, then bubbled with CO2 until saturation and plugged with a Bunsen valve. The tubes were placed later on in a tube rack and incubated to 39° C during 48 h. All the samples were vigorously shaken every 12 h. After the incubation period, to each tube was added 1.5 ml of acid pepsin avoiding formation of foam and later were added 3.5 ml of the same solution (acid pepsin) and incubated for 48 additional hours. Later on the samples were filtered  (Whatman No 5) by gravity and the residues placed in porcelain crucibles for drying to 50-60° C during 48 h. Finally the dry residues were weighed and the digestibility was calculated following the conventional procedures.   

Chemical analysis of the samples   

The determination of DM, crude protein, crude fat, crude fiber, ashes and  nitrogen free extract were carried out following the conventional procedures (AOAC 1990). The fraction of the non-protein nitrogen (NPN)  was determined by weighing 2 g of sample and placing it in an Erlenmeyer  flask of 125 ml of capacity. To each sample was added 25 ml of water, with shaking for 10 minutes and after a rest of 30 minutes, 25 ml of a 20% solution of  tricloracetic acid was added, followed by shaking for a further 10 minutes and allowed to rest during 3 h. Finally the samples were filtered. Crude protein in the residue was analyzed following the procedure described by Tejada (1992) and the soluble protein calculated by difference.   

Statistical analysis   

The values obtained for the chemical composition of the feeds and their digestibilities were analyzed by “T” test and linear regression using the statistical package Number Cruncher Statistical System 2001 (NCSS, Statistical Software).  

Results and discussion   

The proportion of DM in the vegetable wastes was generally low except for the ear maize byproducts (Table 1). The leaves of cabbage, radish, lettuce and  nopal-legume were promising as sources of protein with values similar to that in sun-cured lucern. The lipid content was low in al the feed sources. 

Table 1. Chemical composition of the feeds utilized in the dairy production systems of Iztapalapa (DM basis except for DM content which is on fresh basis)


Crude Protein

Crude Fat

Crude Fiber


Nitrogen-free Extract

Cabbage leaves


18.9 b



17.1 b

49.1 b

Whole ear of maize

22.2 b


2.3 b



68.3 b

Cauliflower leaf


25.7 b

2.9 b


19.2 b


Maize ear leaf




26.7 b


63.3 b

Radish leaf,


26.8 b

2.2 b


30.7 b




18.1 b

2.2 b


28.5 b



5.7 b

14.7 b

1.8 b


22.7 b

48.8 b

Maize cob

23.4 b



27.9 b


64.9 b

Sun-cured lucerne

14.4 b

27.5 b

2.3 b

24.6 b

12.7 b









Means without letter in common differ at P <0.05 

NPN  values were low in all the vegetable wastes which implies that most of the nitrogen was in the form of protein (Table 2). In vitro DM and OM digestibilities were high for all the samples other than the maize cobs. 

Table 2. Protein fractions and in vitro digestibility of the dry matter (DM) and  organic matter (OM) in the vegetable wastes used in  the dairy production systems in Iztapalapa. 

Soluble Protein

Non Protein Nitrogen

DM Digestibility

OM Digestibility

Cabbage leaves

54 b


82.3 b

99.2 b

Maize cob (with leaves)

61 b

6.2 b

85.2 b

87.9 b

Cauliflower leaves

47 b

1.8 b

79.0 b

97.6 b

Maize ear leaves



77.7 b

79.8 b

Radish leaves

58 b

2.5 b


88.0 b




71.2 b

99.1 b






Maize cob

57 b




Sun-cured lucerne

69 b

3.0 b

75.1 b

86.1 b






Means without letter in common differ at P <0.05 

As expected the OM digestibility showed a positive relationship with the content of crude protein and negative with the crude fiber (Table 3).     

Table 3.  Relationships between the contents of crude protein and crude fiber with the digestibility of the organic matter of the feeds



Regression equation

Standard Error



Crude protein


Y=76.2 +0.71x




Crude fiber


Y=112 -1.3x




OMD Organic Matter Digestibility

The results obtained in the study showed that the chemical composition and digestibility of most of the waste feeds used by the producers of milk in the area were similar to those reported in the literature for the conventional forages (Astiasaran and Martínez 2000; Gohl 1982). Possible disadvantages would be the low DM content, and possible negative effects on the DM intake of the milking animals. This latter situation explains why most of the producers give maize stubble and/or maize cobs as a "filler" feed, when in fact the main effect could be an increase in the DM content within the rumen to improve the voluntary intake (Salcedo 1997) at the expense of OM digestibility,  because of the increase in crude fiber (Table 3). We have some evidence (Arias et al 2002) that changes  in the composition of the milk are associated with possible changes in the pattern of rumen fermentation as a result of  lower consumption of DM. 


The authors thank the dairy producers in Iztapalapa for the donation of the samples for the chemical analysis; the authorities of the Metropolitan Autonomous University for the use of their facilities; Mr. Manuel López G for assistance in the mathematical analyses.   


AOAC 1990 Association Of Official Analytical Chemists. Methods of Analysis 15 th Edition Washington D.C.   

Arias L, López M, Losada H, Cortés J, Vieyra J, Soriano R and Grande D 2002. A note on the effect of molasses/urea lick blocks on the production of milk and corporal conformation of dairy cows maintained in the urban environments of Iztapalapa, Mexico City: A case study. Livestock Research for Rural Development. Volume 14 Number 6 

Astiasaran I y Martínez  J A 2000 Alimentos: Composición y Propiedades Edición. McGraw-Hill. Madrid. pp 228-236.  

Grande D, Losada H, Rivera J, Vieyra J and Arias L 1994 Potential of the organic residuals generated in the Central Food Depot of Mexico City for the animal feeding. Memory I International Congress and 2 National Congress of investigation in agricultural production systems. UAM, UAEM. Mexico. pp 339-343.   

Gohl B 1982  Tropical Feeds Informative summaries about feeds and nutritive values. Organization of the United Nations for the Agriculture and the Feeding. Rome. pp 1-218 

Losada H, Bennett R, Vieyra J, Soriano R, Cortes J and Billing S  2000 Recycling of solid wastes in the east of Mexico City by livestock and agricultural production systems. Internet conference Material on Flow Analysis of Integrated Bio-Systems. 

NCSS 2001  NCSS Statistical Software. Keysville, UTA 84037. USA.   

Ruiz R 1995 Digestibility in vitro of forages. Thesis in option of the degree of PhD. University of Reading, England. pp 80-83  

Salcedo G 1997 Determination of the ingestion capacity, digestibility, production and quality of the milk of two vegetative cycles of Sorgo X Sudán for cows Frisonas in metabolic area. Revista de la Facultad de Agronomía (LIGHT), 14:141-152.   

Tejada I l992 Control of quality and analysis of foods for animals. Systems of Continuous Education in Animal Production B.C. Mexico, DF. pp 285-300   

Tilley J M A and Terry R A 1963 The two stage technique for the in vitro digestion of forage crops. Journal of British Grassland Society 18:104-109   


Received 29 July 2002; Accepted 8 November 2002

Go to top