Wireless sensor networks (WSNs) are among the most promising technologies of the new millennium. The opportunities afforded by being able to program networks of small, lightweight, low-power, computation- and bandwidth-limited nodes have attracted a large community of researchers and developers. However, the unique set of capabilities offered by the technology produces an exciting but complex design space, which is often difficult to negotiate in an application context. Deploying sensing physical environments produces its own set of challenges, and can push systems into failure modes, thus revealing problems that can be difficult to discover or reproduce in simulation or the laboratory. Sustained efforts in the area of wireless networked sensing over the last 15 years have resulted in a large number of theoretical developments, substantial practical achievements, and a wealth of lessons for the future. It is clear that in order to bridge the gap between (on the one hand) visions of very large scale, autonomous, randomly deployed networks and (on the other) the actual performance of fielded systems, we need to view deployment as an essential component in the process of developing sensor networks: a process that includes hardware and software solutions that serve specific applications and end-user needs. Incorporating deployment into the design process reveals a new and different set of requirements and considerations, whose solutions require innovative thinking, multidisciplinary teams and strong involvement from end-user communities. This special feature uncovers and documents some of the hurdles encountered and solutions offered by experimental scientists when deploying and evaluating wireless sensor networks in situ , in a variety of well specified application scenarios. The papers specifically address issues of generic importance for WSN system designers: (i) data quality, (ii) communications availability and quality, (iii) alternative, low-energy sensing modalities and (iv) system solutions with high end-user added value and cost benefits. The common thread is deployment and deployment evaluation. In particular, satisfaction of application requirements , involvement of the end-user in the design and deployment process, satisfactory system performance and user acceptance are concerns addressed in many of the contributions. The contributions form a valuable set, which help to identify the priorities for research in this burgeoning area: * Robust , reliable and efficient data collection in embedded wireless multi-hop networks are essential elements in creating a true deploy-and-forget user experience. * Maintaining full connectivity within a WSN, in a real world environment populated by other WSNs, WiFi networks or Bluetooth devices that constitute sources of interference is a key element in any application, but more so for those that are safety-critical, such as disaster response. * Awareness of the effects of wireless channel, physical position and line-of-sight on received signal strength in real-world, outdoor environments will shape the design of many outdoor applications. Thus, the quantification of such effects is valuable knowledge for designers. * Sensors' failure detection, scalability and commercialization are common challenges in many long-term monitoring applications; transferable solutions are evidenced here in the context of pollutant detection and water quality. * Innovative, alternative thinking is often needed to achieve the desired long-lived networks when power -hungry sensors are foreseen components; in some instances, the very problems of wireless technology, such as RF irregularity, can be transformed into advantages. * The importance of an iterative design and evaluation methodology —from analysis to simulation to real-life deployment—should be well understood by all WSN developers. The value of this is highlighted in the context of a challenging WPAN video-surveillance application based on a novel Nomadic Access Mechanism. * Cost benefits to be drawn from devising a WSN based solution to classic application areas such as surveillance are often a prime motivator for WSN designers; an example is offered here based on the use of intelligent agents for intrusion monitoring. * Last but not least, the practicality and usability of the WSN solutions found for novel applications is key to their adoption. This is particularly true when the end-users of the developed technology are medical patients. The importance of feedback, elegant hardware encapsulation and extraction of meaning from data is presented in the context of novel orthopedic rehabilitation aids. Overall, this feature offers wide coverage of most issues encountered in the process of design, implementation and evaluation of deployable WSN systems. We trust that designers and developers of WSN systems will find much work of value, ranging from lessons learned, through solutions to known hurdles, to novel developments that enhance applications. Finally, we would like to thank all authors for their valuable contributions!