Nitrogen (N) fertiliser is one of the main inputs of intensive wheat cropping systems in the UK. The average application of N fertiliser in the UK for winter wheat was 220 kg N ha-1 in 2011, but typically only 50% of this applied N is taken up by the crop. Breeding crops for high N use efficiency (NUE; grain dry matter yield per unit availability of N from soil and/or fertiliser) while maintaining acceptable yield, is widely accepted as one of the approaches to maximise farmers’ revenue and minimise pollution risk by reducing N fertiliser application. NUE is divided into two components; N uptake efficiency (NUpE; crop N uptake per unit availability of N from soil and/or fertiliser) and N utilisation efficiency (NUtE; grain dry matter yield per crop N uptake). NUE can be improved by improving NUpE and /or NUtE. Targeted integration of favourable traits into breeding programmes needs adequate diversity of the existing genotypes. However, genetic diversity of modern bread wheat is narrowed due to chance hybridisation events during wheat evolution and selective breeding for high yield. Hence, it has become necessary to search for novel sources of genetic variation for NUE. One of the sources could be the ancient wheat species, which provide novel resources of genes to improve NUE of modern bread wheat. The overall objective of the current project is to explore favourable physiological traits for NUE in ancient wheat species. Three field experiments, four glasshouse experiments and three growth room experiments were conducted at Sutton Bonington Campus, University of Nottingham. The current work can be divided into four main components; (i) investigation of general plant growth and development, (ii) quantification of NUE and its components, (iii) exploration of root architectural traits of seedlings and mature plants, and (iv) identification of seedling root QTLs related to NUpE in a wheat x spelt population. Significant variation for plant establishment and development was identified among wheat species where bread wheat had high plant establishment and faster plant development compared to ancient wheat species. Above ground biomass production of bread wheat, spelt and emmer was either similar or bread wheat was slightly higher at maturity under high N levels (HN). However, high above ground biomass production at zero N (no N fertiliser; NN) was observed in spelt genotypes. High grain yield of bread wheat is due to improved harvest index when compared to tall ancient wheat species. Delayed onset of flag leaf senescence, slow senescence rate and prolonged leaf greenness were observed in some spelt genotypes, especially spelt cv. Oberkulmer and were positively associated with biomass production. Flag leaf length, width, green area, specific leaf area and SPAD values were significantly different between genotypes used in the study. The greatest green area, SPAD value and maximum width were recorded in bread wheat. NUE was higher in bread wheat followed by spelt, emmer and einkorn. Emmer 2 had higher NUpE than spelt, bread wheat and einkorn genotypes. Delayed senescence hence extended green area duration and deep and vigorous root system might have influenced NUpE of spelt genotypes. High fertiliser N recovery rate of emmer 2 may be associated with well distributed, shallow, horizontally grown root system, especially at early plant growth. NUtE of bread wheat is higher than ancient wheat species and closely related to harvest index and reduced plant height. NUtE is more controlled by genetic factors than NUpE. N supply has a negative relationship with NUpE and its components. Relationships between NUE and its components were not consistent between experiments in the study. However, NUpE and NUtE explained more of the variation in NUE under HN than NN. Root system architecture varied significantly between wheat species used in the study. Seminal root characters of the seedling such as tip angle, number of seminal roots, seminal root length and total root length are closely related to mature root systems. Spelt had long seminal roots with narrow tip angles and hence develops deep root system while having a well-developed superficial root system due to nodal roots. Therefore, both top soil scavenging and deep soil foraging occurs efficiently. A large and deep root system contributes significantly to greater above-ground biomass and green area production of spelt. A greater number of seminal roots with wide tip angles help emmer to develop a root system architecture well adapted to take up fertiliser N from the top soil layers, especially in early stage of the plant growth. There is a possibility of using seedling root traits such as total root length and average length of the seminal root to predict NUpE at maturity. Therefore, these traits might be used as selection criteria for crop breeding for efficient N uptake. Emmer had favourable seedling root traits related to NUpE while spelt showed more favourable root traits at for N uptake at maturity than modern bread wheat and einkorn. Root N uptake efficiency and specific absorption rate of N were higher in emmer than spelt, bread wheat and einkorn. Phenotypic evaluation and genetic analysis of recombinant inbred lines (RILs) produced from the cross between Swiss winter wheat variety Forno (Triticum aestivum) and Swiss winter spelt cv. Oberkulmer (T. spelta) was carried out. Two parents of the population were significantly different for a number of traits, including: number of seminal roots, average length of seminal roots, total root length, maximum width, width to depth ratio, tip angle and emergence angle of seminal root. All measured root traits of seedlings of the RILs varied significantly. A total of 26 significant QTLs were identified for seedling root traits. These QTLs were located on ten different chromosomes; 1BS, 2A, 2D, 3A, 3B, 4A, 5A, 5B, 7AL and 7D. QTL coincidence was found for total root length either with number of seminal roots or seminal root length of the seedlings which showed strong phenotypic relationship. RILs F5-10, F5-36, F5-134, F5 146, F5-230 and F5-234 could be efficient genotypes for N uptake at maturity. The contribution of the A genome to the phenotypic variation observed within this cross for the development of the seedling root system is more important than the B and D genomes, as assessed by QTL analysis. Further studies are needed to identify QTLs associated with NUpE and root architectural traits of emmer and spelt. Favourable root traits in emmer and spelt related to NUpE could be introduced to bread wheat through direct crossing or creating synthetic wheat. The introgression of spelt into modern bread wheat will be useful for molecular studies to develop marker assisted breeding for high NUpE.